At X2, we’re scaling support to over 29 million people using our clinically backed AI chatbot, Tess

X2AI

x2ai

How Technology Can Help with Mental Health Diagnosis

Tell me, how does this sound: local weather updates, info on your flight, the latest news and much more all available in one place. In one place on your phone. Pretty cool right? Well, the coolness increases as there are no humans involved. Only you and a program that can understand and respond to certain phrases. These, my friend, are chatbots. In today’s world, the best way for a company to reach its consumer is through social platforms. According to an Oracle infographic, 90% of businessesuse Facebook to respond to service requests. As Deep Learning and NLP advanced so did the prospect of having a bot that would talk to the consumers. This same advancement has helped us apply automation in every place possible. So why not automate therapy as well? Mean number of mentally unhealthy days during past 30 days among adults aged ≥18 years Given the rise in mental illness in the United States and Worldwide and the lack of mental health infrastructure in some countries. We need to ask, Can chatbots provide a temporary (or permanent) form of therapy? Does it sound bizarre? Well maybe not. The Past In 1950’s Alan Turing published his now famous paper, ‘Computing Machinery and Intelligence’. To measure a machine’s intelligence, he designed the Turing test. If a machine can fool a human, in a conversation, to believe it to be human then it passes the test. In 1966, Joseph Weizenbaum, a professor at MIT Artificial Intelligence Laboratory, took interest in this. He made Eliza, the first chatbot. It was designed to fool people into believing that they were talking with a therapist rather than a bot. It worked by recognizing words from the input to reproduce a response. Then responded with certain keywords from a list of pre-programmed responses. Although with our current understanding of language we can build more complicated conversation interfaces, Eliza is still impressive. Other chatbots such as PARRY and JABBERWACKY have come along, but these weren’t made to simulate therapists. The Present However, we have come a long way from Eliza. Here are some of today’s therapy bots: , a New Zealand based company that made Nadia. Nadia is an animated human female that interacts with the user via a webcam. Actress Cate Blanchett has voiced Nadia and Dr. Mark Sagar has designed her. Her emotional intelligence, helps her look at the user’s face and connect with them on an emotional level. Although, she is a chatbot she is not a therapist. Nadia is there to support and help patients navigate the Australian NDIS(National Disability Insurance Agency). You can see her in action . But, the government has stalled her development due to certain risks related to high-tech projects in the past. Nadia: Soul Machines here Not a chatbot, but a worthy mention since this article is about mental health. Therachat is an application designed for psychotherapist and client communication. It has various features like assigning homework, emotion tracking, scheduling and direct messaging. Although, chatbots are on the rise Therachat’s ambition to enhance the therapist experience is also valuable. Therachat : Made by USC’s Institute for Creative Technologies, Ellie is a bot like Nadia. Funded by DARPA, she is part of a virtual reality program called SimSensei. She has a webcam and a sensor which she uses to read emotion and communicate. Her users are veterans with depression or PTSD (post-traumatic stress disorder). People are less open to revealing personal information to a human as compared to a bot, hence her purpose. And as this , it seems to work! But she warns her users that she isn’t a therapist but here to talk. Ellie: report reads : Built by Microsoft and hosted by Weibo, this chatbot is closer to a best-friend than a therapist. Microsoft built her by mining the Chinese internet for human conversations. She even uses emojis! Compared to Tay, whose racism got it suspended, Xiaoice has a much calmer personality. The Chinese user is open to talking to it about their breakups, family problems and anything else. It saves certain data from past conversations so it can bring them up in the future. The latter part has users doubting if they can trust it with their personal information. Given the amount of internet censorship in China, it is not surprising that she avoids sensitive political questions. Xiaoice An example conversation This isn’t a chatbot, but a startup that makes psychotherapeutic chatbots. They have made bots like Tess, “a psychological AI that administers highly personalized psychotherapy, psycho-education, and health-related reminders, on-demand”. Other include Emma, a Dutch-language bot designed to help people with mild anxiety. And Karim, a bot built to counsel Syrian refugees. This is clever. The cost of building a bot is far lower than sending a group of psychotherapists to the war zone. Also, since people have to open up to a bot they aren’t worried about social stigma. But, we can discuss it’s effectiveness for another time. X2AI : : I left the best for last. Woebot has a lot going for it apart from Andrew Ng coming as a part of its board of directors. You can read his announcement . Woebot isn’t a licensed therapist but he’s being used to target depression. The reason is the same: a growing mental disorder and stigma behind it. Woebot tries to understand what the user is saying and presents them with fun videos and good advice (try it!). Woebot here The Future It isn’t hard to notice the advantages of a therapeutic chatbot. When we don’t talk to a human, we are more likely to open up about our problems. These bots can never replace psychotherapists but they can improve the experience. As we saw with Therachat, helping the therapist in treating the patient is itself a big step. A mental health crisis and lack of infrastructure might make bots our saviors. We can have a world where everyone has their personal therapist. Like Xiaoice, they can talk to it without fear of being judged or ostracized. In extreme cases, the bot can contact family/spouse or send them to a proper psychotherapist. We shouldn’t hope for an all-around general therapist. Instead, what we might end up getting is a bunch of bots that help with different disorders. Take games for example. AI’s like DeepBlue and AlphaGo are far better at playing games than humans. But the same human that plays Chess can also play Go. As humans, we can combine all these things and do them and yet an AI cannot. Another example: a household bot. We dreamt of having a general purpose bot. A bot that could clean everything from clothes to dishes to the carpet. What we instead got was a Roomba, a dishwasher, and a washing machine. Perhaps that is where the future of psychotherapy lies. In different bots helping millions, open up about their issues and improving their lives.

The Chatbot will see you now or will it?

Hallucinations occur when an AI model provides a completely fabricated answer, believing it to be a true fact. The model is convinced that it has produced the correct answer (and is confident in doing so), yet the answer it provides is inherently nonsensical. We observed this behavior in ChatGPT, but it is a phenomenon that can actually occur with all AI models, where the model asserts a confident prediction that ultimately proves to be inaccurate. The most effective approach to addressing this issue is to comprehend our models and their decision-making processes, a concept that is covered by a field I am excited to discuss – explainable AI (XAI)! Watch the ... video https://youtu.be/6QVMCKmRJjQ?embedable=true&transcript=true Now that we have had an introduction to explainability in AI, we can dive deep into the topic of AI ethics and discuss why explainability is such a crucial aspect, provided by my friend Auxane Boch, an expert in the field, a research associate, and a . TUM IEAI freelancer At its core, explainability means that we understand how and why an AI system makes its decisions. This is important for several reasons. Firstly, without understanding the reasoning behind a decision, we can't blindly trust the system's output. This is particularly important in fields like healthcare or finance, where a wrong decision can have significant consequences. Secondly, without explainability, we can't identify and correct any biases or errors in decisions that may exist in the system. This is vital to ensure fairness, inclusivity, and the robustness of a system! However, there are some major challenges to achieving explainability in AI systems. When we talk about explainability, we can see it from two angles: technical transparency - usually targeted at engineers or technical experts, and understandability - for all types of users, from a tech company CEO to your baker. As you can imagine, understandability is a big challenge in itself! You cannot explain things in the same way to a highly digitally literate person as you would to your grandparents. Thus, one of the biggest challenges is the variation in digital literacy. Not everyone has the technical background to understand the complex algorithms and processes behind AI. This means that even if an explanation is provided, it may not be easily comprehensible to everyone. This could lead to a lack of trust in the system or even skepticism towards AI as a whole. In turn, this skepticism will impact the adoption of technologies and their acceptability. Another challenge is the diversity of cultures we have the chance to see in our world. Different cultures may have different expectations regarding explainability and ways to understand - just as we know counting can be different depending on the culture. For example, some cultures may value transparency and a clear understanding of decision-making processes through visualization and a lot of details, while others may prioritize accurate outcomes over explanations. This means that AI developers must be aware of cultural differences and adapt their systems according to their target population. In conclusion, explainability is a crucial aspect of AI ethics that enables us to trust and understand AI systems. However, achieving explainability is not without its challenges, including digital literacy and cultural differences. As we continue to develop and implement AI systems, we must strive to achieve an acceptable level of explainability while being aware of these challenges and working to overcome them. I hope you've enjoyed this ethics-related piece and the introduction to XAI! Auxane shares many such short pieces in if that is something you find interesting as well! my weekly newsletter

Tackling AI Hallucinations: The Importance of Explainability in AI (XAI)

An inmate at a New York correctional facility, Glenn Rodriguez, was due for parole soon. The man has been on his best behavior and was looking forward to being released and starting a new life. To Glenn’s horror, he was denied parole. An AI algorithm the parole board used gave this inmate a poor score, and as it wasn’t explainable, no one knew that something was going terribly wrong. Mr. Rodriguez fought his case and eventually was released after spending another unnecessary year in prison. Unfortunately, this type of mistake can occur whenever AI is deployed. If we don’t see the reasoning behind algorithms’ decisions, we can’t spot the problem. You can prevent this issue in your organization by following the explainable AI principles while . developing your artificial intelligence solution So, what is explainable artificial intelligence (XAI)? How to decide on the right level of explainability for your sector? And which challenges to expect on the way? What is explainable AI, and why should you care? When speaking of AI, many people think of black-box algorithms that take millions of input data points, work their magic, and deliver unexplainable results that users are supposed to trust. This kind of model is created directly from data, and not even its engineers can explain its outcome. Black-box models, such as neural networks, possess superior skills when it comes to challenging prediction tasks. They produce results of remarkable accuracy, but no one can understand how algorithms arrived at their predictions. In contrast, users can understand the rationale behind its decisions with explainable white box AI, making it increasingly popular in business settings. These models are not as technically impressive as black-box algorithms. Still, their transparency is a tradeoff as it offers a higher level of reliability and is preferable in highly regulated industries. What is explainable AI? Explainable AI (XAI) refers to a set of techniques, design principles, and processes that help developers/organizations add a layer of transparency to AI algorithms so that they can justify their predictions. XAI can describe AI models, their expected impact, and potential biases. Human experts can understand the resulting predictions and build trust and confidence in the results with this technology. When speaking of explainability, it all boils down to what you want to explain. There are two possibilities: Explaining AI model’s pedigree: how the model was trained, which data was used, which types of bias are possible, and how they can be mitigated. Explaining the overall model: this is also called “model interpretability.” There are two approaches to this technique: : a more understandable model, such as a decision tree, is used as an approximation of a more cumbersome AI model. Even though this technique gives a simple overview of what to expect, it remains an approximation and can differ from real-life results. Proxy modeling : designing AI models in a way that forces simple easy-to-explain behavior. This technique can result in less powerful models as it eliminates some complicated tools from the developer’s toolkit. Design for interpretability Source Explainable AI principles The US National Institute of Standards and Technology (NIST) developed : four explainable AI principles The system should be able to explain its output and provide supporting evidence (at least). There are several explanation types: ○ Explanations that benefit the end-user ○ Explanations that are designed to gain trust in the system ○ Explanations that are expected to meet regulatory requirements ○ Explanations that can help with algorithm development and maintenance ○ Explanations that benefit the model’s owner, such as movie recommendation engines The given explanation has to be meaningful, enabling users to complete their tasks. If there is a range of users with diverse skill sets, the system needs to provide several explanations catering to the available user groups This explanation needs to be clear and accurate, which is different from output accuracy The system has to operate within its designed knowledge limits to ensure a reasonable outcome Explainable AI example XAI can provide a detailed model-level explanation of why a particular decision was made. This explanation comes in a set of understandable rules. Following a simplified loan application example below, when applicants are denied a loan, they will receive a straightforward justification: everyone over 40 years old saves less than $433 per month and applies for credit with a payback period of over 38 years will be denied a loan. The same goes for younger applicants who save less than $657 per month. Why is explainable AI important? In some industries, an explanation is necessary for AI algorithms to be accepted. This can be either due to regulations and/or human factors. Think about brain tumor classification. No doctor will be comfortable preparing for a surgery solely based on “the algorithm said so.” And what about loan granting? Clients who got their application denied would want to understand why. Yes, there are more tolerant use cases where an explanation is not essential. For instance, predictive maintenance applications are not a matter of life or death, but even then, employees would feel more confident knowing why particular equipment might need preemptive repair. Senior management often understands the value of AI applications, but they also have their concerns. According to Gaurav Deshpande, VP of Marketing at TigerGraph, there is always a “but” in executives’ reasoning: “…but if you can’t explain how you arrived at the answer, I can’t use it. This is because of the risk of bias in the black box AI system that can lead to lawsuits and significant liability and risk to the company brand as well as the balance sheet.” The ideal XAI solution is the one that is reasonably accurate and can explain its results to practitioners, executives, and end-users. Incorporating explainable AI principles into intelligent software: Brings relief to system users. They understand the reasoning and can get behind the decisions being made. For example, a loan officer will be more comfortable informing a customer their loan application was declined if he can understand how the decision was made. Ensures compliance. By verifying the provided explanation, users can see whether the algorithm’s rules are sound and in accordance with the law and ethics. Allows for system optimization. When designers and developers see the explanation, they can spot what is going wrong and fix it. Eliminates bias. When users view the explanation, they can spot any presence of biased judgment, override the system’s decision, and correct the algorithm to avoid similar scenarios in the future. Empowers employees to act upon the system’s output. For example, an XAI might predict that a particular corporate customer will not renew their software license. The first reaction of the manager can be to offer a discount. But what if the reason behind leaving was poor customer service? The system will tell this in its explanation. Empowers people to take action. XAI enables the parties affected by certain decisions to challenge and potentially change the outcome (such as mortgage granting situations). Which industries need XAI the most? Healthcare Finances Automotive Manufacturing 1. Explainable AI in healthcare AI has many . Various AI-powered medical solutions can save doctors’ time on repetitive tasks, allowing them to primarily focus on patient-facing care. Additionally, algorithms are good at diagnosing various health conditions as they can be trained to spot minor details that escape the human eye. However, when doctors cannot explain the outcome, they are hesitant to use this technology and act on its recommendations. applications in healthcare One example comes from Duke University Hospital. A team of researchers installed a machine learning application called , which would send an alert when a patient was at risk of developing sepsis. The researchers discovered that doctors were skeptical of the algorithm and reluctant to act on its warnings because they did not understand it. Sepsis Watch This lack of trust is passed to patients who are hesitant to be examined by AI. Harvard Business Review published a study where participants were invited to take a free assessment of their stress level. registered for the test when they knew a human doctor would do the evaluation. Only 26% signed up when an algorithm was performing the diagnosis. 40% of the participants When it comes to diagnosing and treatments, the decisions made can be life-changing. No surprise that doctors are desperate for transparency. Luckily, with explainable AI, this becomes a reality. For example, Keith Collins, CIO of SAS, mentioned his company is already developing such a technology. Here is what he said: “We’re presently working on a case where physicians use AI analytics to help detect cancerous lesions more accurately. The technology acts as the physician’s ‘virtual assistant,’ and it explains how each variable in an MRI image, for example, contributes to the technology identifying suspicious areas as probable for cancer while other suspicious areas are not.” 2. XAI in finances Finance is another heavily regulated industry where decisions need to be explained. It is vital that AI-powered solutions are auditable; otherwise, they will struggle to enter the market. AI can help assign credit scores, assess insurance claims, and optimize investment portfolios, among other applications. However, if the algorithms provide biased output, it can result in reputational loss and even lawsuits. Not long ago, with its Apple Card product, which was inherently biased against women, lowering their credit limits. Apple’s co-founder, Steve Wozniak, confirmed this claim. He recalled that together with his wife, they have no separate bank accounts nor separate assets, and still when applying for Apple Card, his granted limit was ten times higher than his wife’s. As a result of this unfortunate event, the company was investigated by the New York State Department of Financial Services. Apple made headlines With explainable AI, one can avoid such scandalous situations by justifying the output. For example, loan granting is one use case that can benefit from XAI. The system would be able to justify its final recommendation and give clients a detailed explanation if their loan application was declined. This allows users to improve their credit profiles and reapply later. 3. Explainable AI in the automotive industry Autonomous vehicles operate on vast amounts of data, requiring AI to analyze and make sense of it all. However, the system’s decisions need to be transparent for drivers, technologists, authorities, and insurance companies in case of any incidents. Also, it is crucial to understand how vehicles will behave in case of an emergency. Here is how Paul Appleby, former CEO of a data management software company Kinetica, voiced his concern: “If a self-driving car finds itself in a position where an accident is inevitable, what measures should it take? Prioritize the protection of the driver and put pedestrians in grave danger? Avoid pedestrians while putting the passengers’ safety at risk?” These are tough questions to answer, and people would disagree on how to handle such situations. But it is important to set guidelines that the algorithm can follow in such cases. This will help passengers decide whether they are comfortable traveling in a car designed to make certain decisions. Additionally, after an incident, the provided explanation will help developers improve the algorithm in the future. 4. Explainable artificial intelligence in manufacturing AI has many applications in manufacturing, including predictive maintenance, inventory management, and logistics optimization. With its analytical capabilities, this technology can add to the “tribal knowledge” of human employees. But it is easier to adopt decisions when you understand the logic behind them. Heena Purohit, Senior Product Manager for IBM Watson IoT, explains how their AI-based maintenance product approaches explainable AI. The system offers human employees several options on how to repair a piece of equipment. Every option includes a percentage confidence interval. So, the user can still consult their “tribal knowledge” and expertise when making a choice. Also, each recommendation can project the knowledge graph output together with the input used in the training phase. Challenges on the way to explainable AI The need to compromise on the predictive power Black box algorithms, such as neural networks, have high predictive power but offer no output justification. As a result, users need to blindly trust the system, which can be challenging in certain circumstances. White box AI offers the much-needed explainability, but its algorithms need to remain simple, compromising on predictive power. For example, where algorithms produce remarkable results classifying brain tumors and spotting breast cancer faster than humans. However, when doctors decide on patients’ treatment, it can be a life and death situation, and they want to understand why the algorithm came up with this diagnosis. It can be daunting for doctors to rely on something they do not understand. AI has applications in radiology The concept of explainability There is no universal definition of explainability. It is often a subjective concept. Users might expect one type of explanation, while developers are willing to provide something else. Also, different audiences require tailored justifications, which results in one XAI system having to explain the same output in several different ways. Security and robustness-related issues With XAI, if clients gain access to the algorithm’s decision-making process, they might apply adversarial behaviors, meaning they will take deliberate actions to change their behavior to influence the output. One published a concern that someone with technical skills can recover parts of the dataset used for algorithm training after seeing the explanation, thereby violating privacy regulations. study How to get started with responsible and explainable AI When your company is preparing to deploy responsible XAI solutions, the first thing would be to determine what exactly you need to explain and to whom. Some of the questions addressed during the planning stage can be: Is it a particular decision or the overall model? Who is your audience? Are they technical users? Do you need to provide different explanations for every output, or does one explanation suffice for everyone involved? Next, your company needs to decide on the degree of explainability. Not all AI-based tools require the same degree of interpretability. six application criticality components that will help you determine your desired XAI level: PwC identifies Decision impact: how will the decision affect all the involved parties? This is not limited to revenue-related aspects. For example, an incorrect cancer diagnosis will have a dire impact, while recommending an irrelevant movie will not be a big issue. Control level: is it an autonomous AI system that makes decisions and acts upon them, or does the system’s output recommend actions to humans who can choose whether to follow or not? Risks: any potential harm that the algorithms can cause, such as impacting operations, sales, employees, any ethical, legal, environmental, or health-related considerations. Regulations: the legal framework of the industry and whether the decisions need to be explained. Reputation: how the system interacts with stakeholders — within the business and with the society in general. Rigor: if the application is accurate and able to generalize its conclusions to unseen data. Finally, even after explainable artificial intelligence is in place, it is best to take action to ensure your data usage is ethical. Perform timely audits of your algorithms. A strong explainability feature can reveal any bias sneaking into your software. However, if you settled on a limited explainability solution, bias could remain unnoticed. Also, you can subscribe to the . Or even develop your own set of ethical principles of data usage, as Microsoft did. They included fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability. Partnership on AI consortium On a final note… Implementing traditional , let alone its explainable and responsible version. Despite the obstacles, it will bring relief to your employees, who will be more motivated to act upon the system’s recommendations when they understand the rationale behind it. Moreover, XAI will help you comply with your industry’s regulations and ethical considerations. AI is full of challenges If you have an idea of an explainable AI solution to build, or if you are still unsure of how explainable your software needs to be, . consult ITRex XAI experts

Explainable AI Principles: What Should You Know About XAI

Remote work and work from home are here to stay, . Now that more companies are doubling down on long-term remote, infrastructure, tools and resources are priority. that much is evident What are some of the biggest bottlenecks of virtual corporate culture? Meetings. In 2020, x.ai recorded From Zooms to Meets to phone calls, virtual meetings are growing by the sheer fact that more people are virtual. 182.4% year-over-year growth in the number of meetings scheduled. We can expect to see many advancements in work from home tools and and resources, as seen by the most recent partnership between Newton (previously CloudMagic) and x.ai. The Newton-x.ai partnership allows for seamless calendar scheduling for individuals and teams within the Newton platform. After a user links their x.ai account to Newton, they can share their availability instantly with easy keyboard shortcuts or  even have an AI Scheduler send available times directly to guests, right on the email thread. "The opportunity to pair x.ai scheduling with a user-first email platform was a no brainer. We've worked closely with the Newton team as our first scheduling API partner on what we believe is a simple-to-use and sophisticated way to schedule meetings natively from your inbox," said Dennis R. Mortensen, CEO and founder of x.ai. Founded in 2014, x.ai has spent more than 5 years tuning an AI-powered scheduling engine that instantly generates availability for individuals and teams based on their Meeting Templates. In 2020, x.ai recorded 182.4% year-over-year growth in the number of meetings scheduled on its platform, as scheduling software became a must-have for individuals, small teams, and enterprise companies. Newton itself has a great back story and was acquired by customers and fans Maitrik Kataria, of and Justin Mitchel of , in 2020 after its parent company went bankrupt. Originally one of the first email apps to offer snoozed emails and app integrations they’ve since added new features, fixed long overdue bugs and positioned the once legacy email client as a real contender in the space. Simform SoFriendly.com “It was a surreal moment to go from superuser to owner of my favorite email client. The new normal of work from home has propelled Newton further and faster than I could have hoped -- and has offered everyone the opportunity to become proficient schedulers. We love creating new features that help people get time back and x.ai gave us a great opportunity to do more,” said Maitrik Kataria, co-founder, Newton.

Newton partners with x.ai, now the first native calendar scheduling integration

By: Amanda Mulay Every day, in my role as senior talent manager at , I talk with our founders to learn about their hiring needs and challenges. During these conversations I get to know what they need to grow their teams, learn the ins and outs of the company’s structure, and hear about their stars and emerging leaders. In this new series, “What’s in a role,” I’ll interview individuals holding unique positions across , responsible for propelling crucial initiatives core to the function of the business. Lerer Hippeau our portfolio Kicking off this series is , VP of AI Training at x.ai. From how she got started in AI to her perspective on the future of automation, here’s a look at her day-to-day at , best known as the creator of AI scheduling assistants Amy and Andrew Ingram. Liying Wang x.ai Where did you start your career and how did you end up in your current role at X.ai? Amanda Mulay: I grew up in Palo Alto so tech is part of my constitution. I studied Existentialism and Literary Theory at Cornell, both of which had a profound impact on the way I view the world and have given me the tools to approach this linguistics problem that is one of the biggest challenges within my role. I think having a humanities background has been invaluable to my success in my job today. Liying Wang: I joined Lyft in the early days way back when it was still Zimride, and have been working in tech ever since. The possibilities to change the world in a positive and disruptive way are unlike any other industry. After Lyft, I tried my hand at my own startup, an e-commerce business for sports nutrition, which was an amazing growing experience. Then I joined Facebook in Menlo Park, where I ran the human annotator side of various machine learning products. The breadth of types of products and brilliant product and engineering partners gave me the opportunity to learn how to evaluate, optimize and scale data annotation. I heard about the role at New York-based x.ai through a Crunchbase notification. The company had just raised a Series B and I thought it was a super interesting opportunity and knew I wanted to stay in the AI/machine learning industry. I reached out cold and heard from Dennis [Mortensen] within a day. The rest is history! What does your role entail and how do you typically explain it? Mulay: My job is to solve complex problems in unchartered territory, and I wouldn’t have it any other way! My domain expertise is in building, running, and optimizing our AI Training team. The data annotations produced by this team provide the inputs for our machines to learn how to understand natural language in the production environment. Wang: I designed the infrastructure used by the AI Trainers to annotate data, created the annotation quality controls, processes, feedback mechanisms, and guidelines to teach the AI Trainers how to achieve high quality data annotations. Natural language is fluid and constantly evolving, and all of these components require perpetual flexibility and optimization. At a high level, what happens is this: if the machine has low confidence in its parsing of an email that Amy or Andrew receives, our system sends the AI Trainer team a data annotation task. This data annotation task will then feed back into the machine’s learning so it can handle it with higher confidence in the future. This process happens for specific micro tasks, so humans don’t ever decide how Amy will respond to a given email or get involved in the actual scheduling in any way. The data produced by the AI Training team must align with the machine learning models in addition to the product’s functional requirements. The inextricably linked systems require constant collaboration and fine tuning in order to work in harmony. The way I sometimes explain it to people is to imagine you have a baby robot, and you have to teach the baby robot to perform independent tasks on its own. In order to do that, you need to provide it loads of consistent inputs so it can learn, while accounting for other variables. What are the most common areas of confusion when you’re describing your role in AI? Mulay: I see AI as being able to create opportunity for mankind rather than take opportunity away. Unfortunately the latter is something I hear more frequently than the former. I see this occurring in two general ways. Wang: 1.) Being able to take over a vast variety of tasks which allows people to focus on more challenging, creative ones that AI does not have the cognitive capability of performing. Providing people time to channel their effort and energy towards more intricate and complicated problem-solving is highly beneficial. I certainly see x.ai as providing people increased efficiency, so instead of having to deal with the mundane task of going back and forth to schedule meetings, they can redirect their attention and focus on more important things. 2.) Disrupting industries for the betterment of humanity. AI can (and is starting to) make leaps and bounds towards revolutionizing how we approach other areas such as healthcare or transportation. While most of the misperceptions I hear are about AI itself, in terms of my role, most people are surprised I don’t have an engineering or programming background. The area around AI that is relatively less known is the training part which for me dealing with natural language processing, is basically a linguistics challenge. That’s why my background in Literary Theory, Deconstruction, and Operations align well. What do you find the most challenging about your role or responsibilities? Mulay: What drives me is being able to capture the limitless aspects of language with all of its subtleties/variations/unique eccentricities to fulfill data science and product requirements while achieving a level of quality so high we are striving to hit the ceiling of human performance. Wang: How do you see your role and professional skills growing from here? Mulay: AI is such a dynamic and growing field, so the opportunities are really endless. Whether that’s teaching Amy to learn different types of tasks, adding new features, expanding to other platforms to interact with Amy, or training Amy to understand other languages — all of these opportunities will require me to expand my current role and adjust for the nuances required for each new linguistic challenge. Wang: AI is going to keep growing and changing. It’s an industry that’s rapidly expanding right now and there’s AI everywhere we look. I think about AI as training machines by providing inputs that help teach them to perform tasks on their own. Every industry is trying to become more efficient, so the need for AI is omnipresent. How is this role unique to x.ai? Mulay: The data I’m training is specific to Amy, but the approach, philosophy and tenets are certainly transposable to different needs and products. Wang: What makes you most excited about working this area? Mulay: I love working within a space that is still relatively new. The uncertainty is invigorating, and I love tackling these big hairy problems and having to problem solve and iterate as we go. Wang: Interested in a role at x.ai? Check out current job openings here . Keep up with our community. Subscribe to the and follow us on and . Daily Roundup Twitter Instagram

What’s in a role: Liying Wang, VP of AI Training Operations at x.ai

“Modular” is more than a blockchain buzzword—it’s a fundamental principle that could define the future of web3. The approach, which breaks down blockchain functions into distinct, interconnected layers, is gaining traction across the sector for its ability to enhance scalability, security, and flexibility. From data management to network verification, modular architectures are reshaping how we think about decentralization. This shift isn't just a technical consideration—it has far-reaching implications for transforming users into active network stakeholders, enabling new forms of data monetization, and reshaping the overall trajectory of web3. What’s Modularity in Blockchain Let’s start at the start. Modularity matters in blockchain because, in years gone by, the technology was monolithic. This meant a single chain attempted to handle all functions like execution, settlement, storage and consensus. While initially effective, this design soon encountered scaling challenges, leading to increased costs, reduced flexibility, and high operational overhead. As someone who’s been working in confidential computation and blockchain for a number of years, I can attest to the significant advantages of modular designs that separate functions into standalone yet interoperable layers. For starters, they enhance scalability by allowing each layer to optimize for its specific function. Security is improved through specialized protocols for each component, easily meeting and beating privacy standards set forth by Europe’s GDPR or California’s CPA. Flexibility, meanwhile, increases as developers can update or replace individual modules without overhauling the entire system. Perhaps more importantly, modularity offers cost-effectiveness and choice. By distributing tasks across specialized layers, resource requirements for each component fall. Users and developers can therefore select the modules that best suit their needs, creating a more customizable ecosystem. All of this matters because it helps solve blockchain’s scalability “trilemma”—the challenge of simultaneously achieving decentralization, security, and scalability. By breaking down blockchain functions into modular components, we're seeing innovative solutions that push the boundaries of what's possible in web3. Paving the Way for Scalable and Flexible Infrastructure In my mind, modularity is the key enabler for a new wave of projects and developments in this space. Take Celestia. As a data availability layer for modular blockchains, it's revolutionizing how Ethereum rollups and Layer 2 solutions operate. By providing a dedicated platform for publishing and verifying transaction data, Celestia is addressing one of the key bottlenecks in blockchain scalability. Its use of data availability sampling allows for lightweight verification, making the system more accessible and efficient. Similarly, EigenDA is leveraging Ethereum's existing infrastructure to create a robust data availability solution. By utilizing restaking, EigenLayer creates a more cost-effective and secure environment for rollups to post their transaction data. This approach not only enhances scalability but also maintains the security guarantees of the Ethereum mainnet. CARV, meanwhile, applies modularity to data management in gaming and AI with distinct layers for identity, storage, computation, execution, and verification. Its identity layer, for example, implements the ERC-7231 standard, unifying web2 and web3 identities for a seamless user experience. This modular approach enhances data handling and privacy but also enables flexible monetization, allowing users to selectively share and potentially profit from their gaming data with advertisers while maintaining control. These projects demonstrate how modularity is enabling innovative solutions to longstanding challenges in blockchain. By breaking down complex systems into interoperable components, they're not just improving scalability and flexibility, but also laying the groundwork for a new data infrastructure in web3. Putting Users At The Forefront Of Web3 Beyond the tech, however, it’s exciting to see modularity reshaping the very essence of user participation in blockchain. How? By elevating users from mere participants to essential stakeholders in a network’s integrity and operation. This transformation was powerfully demonstrated late last year when XAI, a Layer 3 gaming solution on Arbitrum, sold tens of thousands of nodes through a public sale. By operating nodes, users can participate in governance while also having the opportunity to earn rewards. The community and sentinel node operators, in this case, qualify to collectively earn 50% of the project's XAI tokens. It's a win-win scenario that aligns user incentives with the network’s goals, fostering a more engaged and committed community. This node success – echoed by other projects like Aethir and CARV – demonstrates how modularity can offer technical enhancements and also create new avenues for user participation and value creation. This model of user involvement is redefining how we think about blockchain networks, moving beyond the traditional roles of validators and users to create a more collaborative ecosystem. This is yet again why modularity matters. It brings us one step closer to a truly decentralized, efficient, and user-empowered digital world—the very promise that blockchain technology set out to fulfill. As developers, users, and businesses continue to explore and leverage modular blockchain designs, we can expect to see even more innovative solutions that push the boundaries of what’s possible in web3. Watch this space. “Modular” is more than a blockchain buzzword—it’s a fundamental principle that could define the future of web3. The approach, which breaks down blockchain functions into distinct, interconnected layers, is gaining traction across the sector for its ability to enhance scalability, security, and flexibility. From data management to network verification, modular architectures are reshaping how we think about decentralization. This shift isn't just a technical consideration—it has far-reaching implications for transforming users into active network stakeholders, enabling new forms of data monetization, and reshaping the overall trajectory of web3. What’s Modularity in Blockchain Let’s start at the start. Modularity matters in blockchain because, in years gone by, the technology was monolithic. This meant a single chain attempted to handle all functions like execution, settlement, storage and consensus. While initially effective, this design soon encountered scaling challenges, leading to increased costs, reduced flexibility, and high operational overhead. As someone who’s been working in confidential computation and blockchain for a number of years, I can attest to the significant advantages of modular designs that separate functions into standalone yet interoperable layers. For starters, they enhance scalability by allowing each layer to optimize for its specific function. Security is improved through specialized protocols for each component, easily meeting and beating privacy standards set forth by Europe’s GDPR or California’s CPA. Flexibility, meanwhile, increases as developers can update or replace individual modules without overhauling the entire system. Perhaps more importantly, modularity offers cost-effectiveness and choice. By distributing tasks across specialized layers, resource requirements for each component fall. Users and developers can therefore select the modules that best suit their needs, creating a more customizable ecosystem. All of this matters because it helps solve blockchain’s scalability “trilemma”—the challenge of simultaneously achieving decentralization, security, and scalability. By breaking down blockchain functions into modular components, we're seeing innovative solutions that push the boundaries of what's possible in web3. Paving the Way for Scalable and Flexible Infrastructure In my mind, modularity is the key enabler for a new wave of projects and developments in this space. Take Celestia. As a data availability layer for modular blockchains, it's revolutionizing how Ethereum rollups and Layer 2 solutions operate. By providing a dedicated platform for publishing and verifying transaction data, Celestia is addressing one of the key bottlenecks in blockchain scalability. Its use of data availability sampling allows for lightweight verification, making the system more accessible and efficient. Similarly, EigenDA is leveraging Ethereum's existing infrastructure to create a robust data availability solution. By utilizing restaking, EigenLayer creates a more cost-effective and secure environment for rollups to post their transaction data. This approach not only enhances scalability but also maintains the security guarantees of the Ethereum mainnet. CARV, meanwhile, applies modularity to data management in gaming and AI with distinct layers for identity, storage, computation, execution, and verification. Its identity layer, for example, implements the ERC-7231 standard, unifying web2 and web3 identities for a seamless user experience. This modular approach enhances data handling and privacy but also enables flexible monetization, allowing users to selectively share and potentially profit from their gaming data with advertisers while maintaining control. These projects demonstrate how modularity is enabling innovative solutions to longstanding challenges in blockchain. By breaking down complex systems into interoperable components, they're not just improving scalability and flexibility, but also laying the groundwork for a new data infrastructure in web3. Putting Users At The Forefront Of Web3 Beyond the tech, however, it’s exciting to see modularity reshaping the very essence of user participation in blockchain. How? By elevating users from mere participants to essential stakeholders in a network’s integrity and operation. This transformation was powerfully demonstrated late last year when XAI, a Layer 3 gaming solution on Arbitrum, sold tens of thousands of nodes through a public sale . sold tens of thousands of nodes through a public sale sold tens of thousands of nodes through a public sale By operating nodes, users can participate in governance while also having the opportunity to earn rewards. The community and sentinel node operators, in this case, qualify to collectively earn 50% of the project's XAI tokens. It's a win-win scenario that aligns user incentives with the network’s goals, fostering a more engaged and committed community. This node success – echoed by other projects like Aethir and CARV – demonstrates how modularity can offer technical enhancements and also create new avenues for user participation and value creation. This model of user involvement is redefining how we think about blockchain networks, moving beyond the traditional roles of validators and users to create a more collaborative ecosystem. This is yet again why modularity matters. It brings us one step closer to a truly decentralized, efficient, and user-empowered digital world—the very promise that blockchain technology set out to fulfill. As developers, users, and businesses continue to explore and leverage modular blockchain designs, we can expect to see even more innovative solutions that push the boundaries of what’s possible in web3. Watch this space.

Why Modularity Matters: Unlocking the Future of Data Infrastructure in Web3

If you've been a victim of a crypto theft in 2024, you're not alone. A new report from blockchain researchers TRM Labs says that hackers stole more money in the first six months of 2024 when compared to the same period last year. Anecdotally, it is likely that at least one person reading this newsletter has been a victim of that theft. Looking at the numbers, hackers made off with $1.4 billion in crypto in the first half of 2024, compared to $657 million in the comparable period of 2023. In short, double the amount. Most of the damage came from a "small number of large attacks" with the top five hacks and exploits accounting for 70% of the amounts stolen, the report said. Hackers relied on private key and seed phrase compromises to make off with users' crypto and also used smart contract exploits and flash loan attacks to steal assets. The biggest attack in 2024 so far has been the theft of over 4,500 bitcoins at Japanese cryptocurrency exchange DMM Bitcoin in May which resulted in a loss of over $300 million based on the cryptocurrency's value at the time. "While the exact cause of the attack remains unknown, potential vectors include stolen private keys or address poisoning—a tactic wherein attackers send tiny amounts of cryptocurrency to a victim’s wallet to create fake transaction histories, potentially confusing users into sending funds to the wrong address in future transactions," TRM Labs said in its report. Is 2024 the worst year for crypto hacks? Apparently not. According to TRM, the worst year for crypto hacks was 2022. While people did not lose as much crypto to hacks in 2023, this resurgence in 2024 is worrying and TRM said it had not observed any fundamental changes in the security of the cryptocurrency ecosystem that may explain this upward trend. It also said that it had not found any significant differences in attack vectors or in the number of attacks between the first halves of 2023 and 2024. The United Nations previously accused North Korean hackers of crypto-related thefts, alleging that crypto was being used to fund the country's nuclear and missile programs. TRM suggests that organizations implement a multi-layered defense strategy to protect themselves from hacks and exploits by unruly actors (who may or may not be North Korean hackers). If you've been a victim of crypto theft, drop me an email, and let's tell the world your story. Even if you don't want to tell your story, that's fine – just don't fall for the charlatans on the internet who claim they can recover your crypto. They probably can't. It's a Great Time to be an AI Startup 🤖 If you're the founder of an artificial intelligence startup, boy do we have some great news for you! Funding for your startup should not be a problem. At least, according to the data available at Crunchbase. Investments in AI startups more than doubled in the second quarter of 2024 to $24 billion, according to Reuters, which cited Crunchbase data. Five out of six billion-dollar funding rounds went to AI companies, Reuters said. Surprisingly, the list included Elon Musk's xAI, which raised $6 billion during the second quarter of 2024. In case you don't know, Musk announced his own AI startup soon after OpenAI's ChatGPT became all the rage. Musk was a "co-founder" of OpenAI but had a spat with the company which resulted in him leaving the firm. Initially, Musk wanted to name his company TruthGPT because he felt generative AI products like ChatGPT were too woke, but then named it xAI because this was around the same time Twitter was rebranded to X and the Tesla owner is weirdly obsessed with the alphabet. xAI's aim is "to advance our collective understanding of the universe," according to its website and the company has a chatbot called Grok which is supposedly trained on tweet data. It's not very good though. Still, it looks like investors are piling up for Musk's xAI, in hopes that Grok will finally stick. In Other News.. 📰 It's Not Germany Selling Bitcoin. It's One of Its States and It Has No Choice. — via CoinDesk
Humane execs leave company to found AI fact-checking startup — via TechCrunch
Meta will now remove some posts attacking ‘Zionists’ as a form of hate speech — via CNN
Microsoft to offer Apple devices to employees in China, cites absence of Android services — via Reuters
The Washington Post debuts AI chatbot — via Axios
San Francisco’s AI boom can’t stop real estate slide, as office vacancies reach new record — via CNBC And that's a wrap! Don't forget to share this newsletter with your family and friends! See y'all next week. PEACE! ☮️ — Sheharyar Khan, Editor, Business Tech @ HackerNoon *All rankings are current as of Monday. To see how the rankings have changed, please visit HackerNoon's Tech Company Rankings page. Tech, What the Heck!? is a once-weekly newsletter written by HackerNoon editors that combine HackerNoon's proprietary data with news-worthy tech stories from around the internet. Humorous and insightful, the newsletter recaps trending events that are shaping the world of tech. Subscribe here. If you've been a victim of a crypto theft in 2024, you're not alone. A new report from blockchain researchers TRM Labs says that hackers stole more money in the first six months of 2024 when compared to the same period last year. Anecdotally, it is likely that at least one person reading this newsletter has been a victim of that theft. Looking at the numbers, hackers made off with $1.4 billion in crypto in the first half of 2024, compared to $657 million in the comparable period of 2023. In short, double the amount. Most of the damage came from a "small number of large attacks" with the top five hacks and exploits accounting for 70% of the amounts stolen, the report said. Hackers relied on private key and seed phrase compromises to make off with users' crypto and also used smart contract exploits and flash loan attacks to steal assets. private key private key seed phrase seed phrase smart contract exploits smart contract exploits flash loan attacks flash loan attacks The biggest attack in 2024 so far has been the theft of over 4,500 bitcoins at Japanese cryptocurrency exchange DMM Bitcoin in May which resulted in a loss of over $300 million based on the cryptocurrency's value at the time. theft theft "While the exact cause of the attack remains unknown, potential vectors include stolen private keys or address poisoning—a tactic wherein attackers send tiny amounts of cryptocurrency to a victim’s wallet to create fake transaction histories, potentially confusing users into sending funds to the wrong address in future transactions," TRM Labs said in its report . report report Is 2024 the worst year for crypto hacks? Apparently not. According to TRM, the worst year for crypto hacks was 2022. While people did not lose as much crypto to hacks in 2023, this resurgence in 2024 is worrying and TRM said it had not observed any fundamental changes in the security of the cryptocurrency ecosystem that may explain this upward trend. It also said that it had not found any significant differences in attack vectors or in the number of attacks between the first halves of 2023 and 2024. The United Nations previously accused North Korean hackers of crypto-related thefts, alleging that crypto was being used to fund the country's nuclear and missile programs. previously accused previously accused TRM suggests that organizations implement a multi-layered defense strategy to protect themselves from hacks and exploits by unruly actors (who may or may not be North Korean hackers). If you've been a victim of crypto theft, drop me an email , and let's tell the world your story. Even if you don't want to tell your story, that's fine – just don't fall for the charlatans on the internet who claim they can recover your crypto. They probably can't. drop me an email drop me an email It's a Great Time to be an AI Startup 🤖 It's a Great Time to be an AI Startup If you're the founder of an artificial intelligence startup, boy do we have some great news for you! Funding for your startup should not be a problem. At least, according to the data available at Crunchbase. Investments in AI startups more than doubled in the second quarter of 2024 to $24 billion, according to Reuters, which cited Crunchbase data. Five out of six billion-dollar funding rounds went to AI companies, Reuters said. Surprisingly, the list included Elon Musk's xAI, which raised $6 billion during the second quarter of 2024. In case you don't know, Musk announced his own AI startup soon after OpenAI's ChatGPT became all the rage. Musk was a "co-founder" of OpenAI but had a spat with the company which resulted in him leaving the firm. spat spat Initially, Musk wanted to name his company TruthGPT because he felt generative AI products like ChatGPT were too woke , but then named it xAI because this was around the same time Twitter was rebranded to X and the Tesla owner is weirdly obsessed with the alphabet. name name too woke too woke weirdly obsessed weirdly obsessed xAI's aim is "to advance our collective understanding of the universe," according to its website and the company has a chatbot called Grok which is supposedly trained on tweet data. It's not very good though. website website very good very good Still, it looks like investors are piling up for Musk's xAI, in hopes that Grok will finally stick. In Other News.. 📰 In Other News.. 📰 It's Not Germany Selling Bitcoin. It's One of Its States and It Has No Choice. — via CoinDesk Humane execs leave company to found AI fact-checking startup — via TechCrunch Meta will now remove some posts attacking ‘Zionists’ as a form of hate speech — via CNN Microsoft to offer Apple devices to employees in China, cites absence of Android services — via Reuters The Washington Post debuts AI chatbot — via Axios San Francisco’s AI boom can’t stop real estate slide, as office vacancies reach new record — via CNBC It's Not Germany Selling Bitcoin. It's One of Its States and It Has No Choice. — via CoinDesk CoinDesk CoinDesk Humane execs leave company to found AI fact-checking startup — via TechCrunch TechCrunch TechCrunch Meta will now remove some posts attacking ‘Zionists’ as a form of hate speech — via CNN CNN CNN Microsoft to offer Apple devices to employees in China, cites absence of Android services — via Reuters Reuters Reuters The Washington Post debuts AI chatbot — via Axios Axios Axios San Francisco’s AI boom can’t stop real estate slide, as office vacancies reach new record — via CNBC CNBC CNBC And that's a wrap! Don't forget to share this newsletter with your family and friends! See y'all next week. PEACE! ☮️ — Sheharyar Khan , Editor, Business Tech @ HackerNoon Sheharyar Khan Sheharyar Khan *All rankings are current as of Monday. To see how the rankings have changed, please visit HackerNoon's Tech Company Rankings page. *All rankings are current as of Monday. To see how the rankings have changed, please visit HackerNoon's Tech Company Rankings page. *All rankings are current as of Monday. To see how the rankings have changed, please visit HackerNoon's Tech Company Rankings page. Tech Company Rankings Tech Company Rankings Tech, What the Heck!? is a once-weekly newsletter written by HackerNoon editors that combine HackerNoon's proprietary data with news-worthy tech stories from around the internet. Humorous and insightful, the newsletter recaps trending events that are shaping the world of tech. Subscribe here. Tech, What the Heck!? is a once-weekly newsletter written by HackerNoon editors that combine HackerNoon's proprietary data with news-worthy tech stories from around the internet. Humorous and insightful, the newsletter recaps trending events that are shaping the world of tech. Subscribe here. Tech, What the Heck!? is a once-weekly newsletter written by HackerNoon editors that combine HackerNoon's proprietary data with news-worthy tech stories from around the internet. Humorous and insightful, the newsletter recaps trending events that are shaping the world of tech. Subscribe here . here here

Are You a Victim of Crypto Theft?

Diversity and Inclusion in AI: Lessons from Human and AI Collaboration

Swarm Summit 2024 Recap

Effective Anomaly Detection Pipeline for Amazon Reviews: References & Appendix

Portfolio Management: All The Ways AI Is Transforming Modern Asset Strategies

পোর্টফোলিও ম্যানেজমেন্ট: সমস্ত উপায় এআই আধুনিক সম্পদ কৌশলগুলিকে রূপান্তরিত করছে

Portfoliomanagement: Wie KI moderne Anlagestrategien verändert

Gestión de carteras: todas las formas en que la IA está transformando las estrategias de activos modernas

Gestion de portefeuille : toutes les façons dont l'IA transforme les stratégies d'actifs modernes

पोर्टफोलियो प्रबंधन: वे सभी तरीके जिनसे AI आधुनिक परिसंपत्ति रणनीतियों को बदल रहा है

ポートフォリオ管理: AI が現代の資産戦略をどのように変革しているか

포트폴리오 관리: AI가 현대 자산 전략을 변화시키는 모든 방법

Gerenciamento de portfólio: todas as maneiras pelas quais a IA está transformando as estratégias modernas de ativos

Управление портфелем: все способы, которыми ИИ трансформирует современные стратегии управления активами

Portföy Yönetimi: Yapay Zekanın Modern Varlık Stratejilerini Dönüştürdüğü Tüm Yollar

Quản lý danh mục đầu tư: Tất cả các cách AI đang chuyển đổi các chiến lược tài sản hiện đại

投资组合管理：人工智能正在改变现代资产战略

Five years ago, Frank Chen posed a question that has stuck with me every day since. The question was this, “If self-driving cars are 51% safer, are we not morally obligated to adopt them?”.  I’ve posed this question a ton of times over the past five years, and usually, a knee-jerk reaction leads to an interesting debate. What makes this question so great is the knife's edge – it’s not 99% safer, it’s not 70% safer, it’s 51% safer. only To put it into context. The National Highway Safety Administration has that in 2022, there were an estimated 42,795 traffic fatalities. 50% of 42,795 is 21,398 people, and 51% is 21,825 people. reported That means if self-driving cars are 51% safer, using them would save 427 people’s lives every year. That is about 1.5 Boeing 777 airplanes full of passengers. Is saving 427 lives a moral argument for adoption? In the five years I’ve been sharing this question, the answers are never simple. They’re always fraught with “what ifs.” But even if the answers lack clarity, I think the question is incredibly important. In part because it opens a broader — and equally important — debate on the moral imperative of AI adoption across many aspects of our lives and work. Because, after all, avoiding technology that could save lives may be just as ethically problematic as adopting technology too hastily, The Moral Imperative of AI Adoption I've always found the debate around autonomous vehicles to be a perfect microcosm for the broader discourse on AI. If we possess technology that's statistically safer than human-operated vehicles, isn't the moral choice obvious? Consider this: have shown that human drivers had a higher rate of crashes with a meaningful risk of injury than self-driving  (AI-powered) cars. Specifically, human drivers caused 0.24 injuries per million miles (IPMM) and 0.01 fatalities per million miles (FPMM), while self-driving cars caused 0.06 IPMM and 0 FPMM. studies And remember, these numbers aren't just statistics. They represent real lives that could be saved by embracing AI technology. But why stop at autonomous vehicles? The potential for AI to enhance safety, efficiency, and accuracy spans across fields like medicine, public health, food safety, agriculture, cybersecurity, crime prevention, and military science. If AI can diagnose diseases with greater accuracy than human doctors, predict crop failures before they devastate food supplies, or thwart cyber-attacks before they breach our data, don't we have a moral obligation to also utilize those technologies? Of course, these are dramatic examples, but the argument extends beyond life-and-death scenarios. AI's ability to improve our daily quality of life is equally compelling. Whether by simplifying mundane tasks or making information and services more accessible and equitable, AI can end drudgery and enhance our daily quality of life. The moral imperative to adopt AI isn't just about preventing harm or death; it's about whether we have an obligation to contribute to human well-being if we can. The Dilemma of Choice and Safety So, do we choose human-operated vehicles (or human-led processes) knowing they are less safe or efficient than their AI counterparts? Simply because they are more human? Faced with the choice between human-operated systems and AI-enhanced alternatives, my thought is the decision should obviously hinge on safety and efficiency rather than an allegiance to some murky idea of what is "human" or not. Embracing AI doesn't mean disregarding human value or input; rather, it's about acknowledging that what's human isn't inherently superior — and honestly, is often significantly inferior in specific contexts. Now please don’t get out the pitchforks, I’m not joining Team Robot Overlord. I get the anxiety a lot of people feel about the disruption AI is already causing to their jobs and the societal change that is undoubtedly headed our way. I just wonder whether the efficiencies of AI and the quality of life benefits might, in the long run, outweigh the impact of those disruptions. Some of our reluctance to adopt AI is informed by cognitive biases and fears. For a species famed for its adaptability, we humans don’t love change. Cognitive biases play a substantial role in our hesitance to embrace AI. Cognitive biases are psychological patterns that are a holdover from our early years as Homo Sapiens. They are the habits our minds fall into — cognitive shortcuts that can be useful when running from predators but definitely skew our modern perception and judgment. In this case, recognizing and addressing these biases is crucial in moving toward a more rational, ethical approach to AI adoption. Here are a few that I think could be in play, influencing our suspicion, trust, or acceptance of AI technologies. People tend to attribute human characteristics to AI or robots, influencing their trust and expectations. This can lead to unrealistic assumptions about AI systems' capabilities or the attribution of evil intentions. Anthropomorphism Bias: This bias leads individuals to overestimate the probability of events associated with memorable or vivid incidents. Sensationalized media reports about AI failures or successes can be blown out of proportion and disproportionately influence perceptions of AI reliability and safety. Availability Heuristic: What’s everyone else doing? People may seek out or interpret information in a way that confirms their preexisting beliefs or hypotheses about AI. This bias can hinder the objective evaluation of AI technologies and their potential benefits or risks. Confirmation Bias: People don’t want to miss out on beneficial technologies but can lack an understanding of the implications. This bias can overshadow critical evaluation and lead to premature adoption. This bias is related to the Bandwagon Effect (e.g., The tendency to do or believe things because everyone else does - or influencers do. People might trust or distrust AI technologies simply because that seems to be the popular sentiment. FOMO: People prefer to maintain the current state, which leads to resistance against adopting something new like AI, regardless of potential benefits or proven superiority. This bias can slow innovation and the adoption of potentially life-enhancing technologies. Status Quo Bias: This bias makes the pain of losing something feel more potent than the pleasure of gaining something of equal value. For AI, this means the fear of job loss or loss of control overshadows the benefits of safety/efficiency/convenience. Loss Aversion: Overestimation of one's ability to control or understand something. For AI, this means overestimating or underestimating the risks associated with AI. Overconfidence Bias: Numbers are scary! People tend to be biased against algorithms, believing human decision-making is superior, even when evidence suggests otherwise. On the other hand, some might also have an unquestioning trust in AI decisions, ignoring the potential for errors or biases in AI systems. Algorithm Aversion/Trust: Economic Rationality Interesting, right? But the truth is, this is all academic. We may not even get to make this decision in the end.  Companies are already making it. A ton of corporations are barreling ahead with AI integration — mainly because the ROI often speaks louder than ethical debates. Take with its significant shift towards automation. The efficiency and economic benefits are tangible and measurable, and in the face of cold hard cash, moral and social criticisms suddenly feel more academic. Amazon as a prime example, Still, this isn't only about stone-hearted capitalism; it's about survival and adaptation. Businesses are tasked every day with the challenge of balancing technological adoption with ethical and ESG responsibilities. The impact of AI on employment and human well-being can't be an afterthought. For thousands, financial stability and career wellness hinge on these decisions. It’s something a lot of enterprises are grappling with. And here's where the moral imperative question becomes more nuanced. If AI can streamline operations, reduce costs, and even create new opportunities, then aren’t we also morally responsible for exploring these technologies? The trick will be to keep that ethical compass handy and ensure that as we embrace AI's efficiencies, we also safeguard against its potential to unfairly disrupt livelihoods. We are in a Transition Period Either way, we need to watch our footing. We are standing on the precipice of a new era, and one solid push could end us in a freefall. AI is no longer a futuristic fantasy; it's absolutely embedded in our daily lives and work. That’s exciting — and scary as hell. One of the most significant challenges we face is the accessibility or tech gap. AI has the potential to democratize technology, making powerful tools available to a broader audience. However, right now, AI's promise is primarily being seen by those who already have a certain level of access, so there is also the potential that AI will exacerbate existing inequalities rather than alleviate them. It’s a period of adjustment, so it will require patience, education, and proactive measures to ensure that the benefits of AI are widely distributed. We’ve got the potential to level the playing field so that AI's potential can be unlocked for everyone, not just the privileged few. The Conundrum of Cooperation Okay, so it is a paradox:  For AI to function optimally alongside humans, it must be superior to us in certain tasks.  But that VERY superiority threatens to displace human roles, fueling resistance and fear among us mortals. This paradox creates a tough "push-pull" for AI; that’s why we’re seeing such heated debate about morality. I believe the solution may be a suite of emerging design philosophies and technologies aimed at bridging the gap between AI and human cooperation in an ethical way. I’ll list them below. They are worth asking ChatGPT about: ensures that AI systems are developed with human needs and values at their core. Human-Centric AI Design (HCAI): demystifies AI decisions, making them understandable and transparent to humans. Explainable AI (XAI): guide the development and deployment of AI systems in a manner that respects human rights and values. Ethical AI Frameworks: learns from and adapts to human feedback, ensuring a synergistic relationship. Adaptive/Responsive AI: involves end-users in the AI development process, ensuring their needs and concerns are addressed. Participatory Design: emphasizes AI's role in enhancing human abilities rather than replacing them. Augmented Intelligence: builds confidence in AI systems through reliability, safety, and ethical assurances. Trustworthy AI: To Self-Drive Cars, or to Not Self-Drive Cars? In wrapping up, I’ll take a stand. I think adopting AI is a moral imperative. In my view, the potential to save lives, enhance our quality of life, and even address long-standing inequalities is too significant to ignore. However, this doesn't mean we should dive headfirst without consideration.  In my opinion, we need to approach AI with a blend of enthusiasm and caution—staying excited to explore its possibilities but mindful of the ethical, social, and economic impact. Thoughtful consideration, robust ethical frameworks, and stringent governance are the keys to unlocking AI's potential responsibly. I’m still open to debate on the subject. So, I’ll throw the question out to you. Reply here or on my and tell me why I’m wrong — or right. I invite your thoughts and comments on this complex issue. LinkedIn thread Are we ready to embrace AI with the moral seriousness it demands? Are you ready to take your next road trip in a self-driving car?

Are We Morally Obligated to Adopt AI?

আমরা কি নৈতিকভাবে এআই গ্রহণ করতে বাধ্য?

Sind wir moralisch verpflichtet, KI einzuführen?

¿Estamos moralmente obligados a adoptar la IA?

Sommes-nous moralement obligés d’adopter l’IA ?

क्या हम नैतिक रूप से एआई को अपनाने के लिए बाध्य हैं?

私たちには AI を導入する道徳的義務があるのでしょうか?

우리는 AI를 채택해야 할 도덕적 의무가 있습니까?

Somos moralmente obrigados a adotar a IA?

Обязаны ли мы морально принять ИИ?

Ahlaki Olarak Yapay Zekayı Benimsetmek Zorunda mıyız?

Chúng ta có nghĩa vụ về mặt đạo đức phải áp dụng AI?

我们在道义上有义务采用人工智能吗？

AI has taken over social media. you have a tool that can dig deep into audience insights, predict the next big content hit, craft posts and visuals on its own, spot consumer trends, and decode public sentiment. Picture this: There is no doubt to say that AI is the backbone of every social media platform out there. It doesn't only help in analyzing what content pops up in your feed, but also cleaning up comments for a better browsing experience, and suggesting new content and friends you might enjoy. But, it poses a challenge for marketers, finding that spot where content appeals to both humans and AI algorithms. When it comes to what “The Algorithm" wants, we're left guessing. And, All too often, we switch from cheap tricks to increased engagement without creating any lasting value for our audiences. Interestingly, AI can lead to more genuine marketing on social media. It helps us understand the vast amount of data from social media platforms. This way, we can create content that truly matters to our audience. AI analyzes posts, comments, and interactions to let us provide what people want and address their interests better. Plus, AI saves a lot of time and money by automating and scaling these processes. That's why top social media marketers are using AI to: Automate repetitive tasks, saving time and money. Expand social media strategies across different platforms. Create posts and campaigns based on data that actually works. You can do this too, no matter your budget or company size. The rest of this post shows you how. How AI is Being Used in Social Media So how exactly can you use artificial intelligence to take your social media marketing game to a whole new level? Let's look at some of the best ways to do this. AI-Powered Content Curation One of the most challenging aspects of social media marketing is coming up with fresh content ideas that can drive engagement for your brand. Instead of copying your competitors or throwing a bunch of content ideas at the wall and hoping one sticks, AI analyzes your data and creates targeted messages based on what your audience has liked in the past. This way, you can be sure that the content you share will resonate with your social media audience and get the reaction you want. AI also helps in improving the quality of your content in terms of tone and grammar. Note: AI-enabled tools even analyze your social media data and determine which keywords, images, hashtags, and colors have a stronger impact on your audience based on past results. In addition, AI-powered content curation tools automatically create social media posts using a URL or information you provide that suits your brand voice. These tools also share the posts on different social platforms. Did you know that Elon Musk says xAI will open-source Grok this week? Automation of Social Network Management Your business is likely on more than one social media platform. And, while consistently publishing quality content, keeping up with the latest trends/updates on every platform can be quite a hassle. However, with the help of AI-powered social media management tools, it is easy to grow your influence and connect with your audience. These tools also plan out your social media posts in advance, track all your profiles, and generate appropriate responses for consumers after analyzing their messages. In simple form, AI will not only plan your campaigns. but also fill content gaps and keep your followers engaged with your brand voice. This tool can recycle content you've previously published, such as blog posts or infographics, and turn them into upload-ready social media posts with hashtags and relevant links. Advertising Campaigns on Social Networks How to get ahead of your competitors and get the most out of advertising on social networks? Several companies are using paid advertising on social media. If you aim to stand out and increase your campaign's impact, it's time to start creating better, smarter ads. AI-powered tools like Sprinklr help you create compelling, personalized social media ads and monitor and optimize your campaigns to maximize ROAS. Such tools offer recommendations to help your campaigns perform better. From comparing ad performance to planning out strategies, AI reduces your overall advertising costs. The copywriting AI tool analyzes your brand language and creates social media ad copy that will increase engagement and conversions for your business. Ensure that your social media ad copy matches your landing page to maintain consistency and communicate with users. Personalization Using Chatbots More followers and clients mean more money, but it also means more time spent serving them. Luckily for you, highly effective customer support and personalized audience experiences are some of the benefits that help your brand grow with the help of artificial intelligence. Hiring enough customer service staff to provide 24/7 customer support may not be feasible. Investing in AI-powered chatbots will respond to audience messages immediately, regardless of the time of day. You can even combine chatbots with your existing live chat software to provide your audience with 24/7 customer service support. With the help of machine learning, your AI chatbots will be able to provide relevant answers to various problems, questions, or queries that your audience may have. Such tools allow you to deliver engaging, seamless, and personalized customer experiences using bots across a variety of social media platforms. Competitor Analysis and Social Listening It's not enough to just know who your competitors are. If you want to get an advantage over them, you need to analyze their social media strategies, find out their weaknesses, and determine what they think. The manually searching and analyzing competitor pages can take weeks or months of grueling work., And, if you succeed in your analysis, there is a high probability that it will contain errors. But with the help of AI, you can easily track your competitors and analyze their social media strategies, reach, engagement rates, conversion rates, brand sentiment, and more. Various AI-powered competitive analysis tools monitor your competitors' social marketing efforts, from user interactions to posts and campaigns. It helps in knowing which parts require a more effective marketing strategy to achieve success. Artificial intelligence tools track mentions of your business, as well as specific hashtags and keywords related to you on social media. Analyzing this information will help you determine whether the public opinion of your brand is negative, positive, or neutral. You will be able to identify potential problems before they become more serious. Emerging AI Technologies in Social Media AI is about to take social media on a futuristic ride. We're not just talking about minor upgrades here - AI is changing the way we scroll, post, and consume content online. Different technologies like synthetic media, AI-created content, and virtual influencers are expanding the limits of what we can do online. Let's take a look at what insanely innovative experiences are coming next… Deepfakes You've probably seen those mind-bending "deepfake" videos going viral where celebrities or politicians seem to be saying and doing things they never actually did. Now creators can create stunningly realistic footage that pushes the boundaries of imagination thanks to artificial intelligence. Deepfakes also open up a world of exciting opportunities for creative self-expression and next-level storytelling through synthetic media. As the technology continues evolving, we'll see deepfakes enabling Hollywood-level visual effects that bring our wildest visions to life. Virtual Influencers This one's unquestionably innovative - companies are using AI to generate incredibly human-looking virtual influencers to advertise products on social media. We're talking lifelike computer avatars with their names, personalities, and millions of followers! Now, imagine the creative potential here. Brands can craft highly engaging, endlessly customizable digital personas to connect with audiences in new ways. Virtual influencers offer limitless brand ambassadorship and storytelling opportunities. AI Content CreatorsImagine telling an AI to whip up your next social media caption, graphic, or video ad...and having it generate something cleverly crafted in seconds. Using large language models and other AI tools, brands can augment human creativity. For content creators, this AI assistance will prove to be an amazing productivity multiplier and source of innovative inspiration. While not replacing humans, AI can elevate our creative visions to new heights while saving tons of time. Feed Personalization Of course, the biggest AI impact on social is automated feed curation. It further deeply understands our preferences to serve up a perfectly personalized experience tailored to each user's interests. This allows for a more rewarding, endlessly engaging social media journey of discovery and relevance. While some worry about filtered bubbles, smarter AI will also ensure we're exposed to a balanced variety of perspectives selected just for us. Conclusion There's no doubt to say that AI is completely changing the world of social media as we know it. From automated content creation to hyper-personalized feeds to wildly realistic synthetic media, this technology is rapidly transforming how we connect, share, and experience online. Artificial intelligence (AI) is making it easier for social media platforms to sort through vast amounts of data. It helps them spot the latest trends, understand what users like, block harmful content, and do much more. AI has a promising future in this space because it makes things better for users and helps brands cater to them more effectively. In social media marketing, AI is crucial because it allows companies to track how well they're doing and pinpoint potential customers.

How AI is Shaping the Future of Social Media

In today’s world, for many people, conversing with AI has become as routine as discussing one’s coffee preferences with a barista (or it soon will be!). Yet, here lies an irony: . the more we interact with AI, the more elusive our understanding of these conversations becomes This irony essentially represents a modern twist on the “black box” dilemma, which has perplexed the ML community for years. The “black box” problem refers to the opaque decision-making processes of ML models (large language models (LLMs) including), where the rationale behind any given response is shrouded in complexity. Despite advances in technology, the inner workings of these models, governed by billions or soon trillions of parameters, remain largely inscrutable. Their decision-making is a puzzle, complicated by nonlinear interactions that defy straightforward interpretation. Prompting — the buzzing word of 2023 — doesn’t make it any better: we now have obscured layers of communication activated with every prompt. What we see — the prompt we type — is merely the surface. Beneath lies a hidden dialogue, an augmented system prompt, which is a complex, coded conversation the model conducts with itself, away from our understanding. And who knows what a model whispers to itself? So, if you were confused about prompting amid the avalanche of articles, blogs, and tutorials about it — you should be. As Ethan Mollick’s reveals, contrary to intuition, , such as pretending to navigate a Star Trek episode or a political thriller, demonstrating that traditional logical or direct prompts may not always yield the best responses from AI. research the most effective prompts involve imaginative scenarios But it also reveals that it’s not coherent and might change with a new version of the model. He mentions the futility of seeking a universal “magic phrase” for AI interaction, the effectiveness of specific prompting techniques like adding context, few-shot learning, and Chain of Thought, and the significant impact that prompts can have on AI performance. But for me — and I’ve been using AI a lot — many times, the most straightforward prompts, or “magic words,” can be surprisingly effective. How to explain it? A few years back, Explainable AI (XAI) was heralded as a solution to the “black box” issue, with entities like DARPA leading the charge (they created the , which has not been updated since 2021). However, the buzz around XAI seems to have dimmed, overtaken by a broader focus on Responsible AI. Is Responsible AI the solution? XAI toolkit So, that’s what we end up with: How do machines make decisions? — We don’t know! How to talk (prompt) to them? — We don’t know as well! But, please, keep shipping to us new, larger (though we will also take smaller) models! Why? — We don’t know! But we can’t stop. Great Article Covering the Topic An explanatory article about the concept of a black box: https://hackernoon.com/deep-learning-and-artificial-neural-networks-solving-the-black-box-mystery-rl7h32wa?embedable=true 🎁 Bonus: The Freshest Research Papers, Categorized for Your Convenience Special Category: Definitely Worth Reading : Discusses the development and advantages of 1-bit LLMs, promising significant cost reductions and efficiency improvements. The Era of 1-bit LLMs Read the paper : Introduces bGPT, a model that simulates the digital world beyond traditional modalities, predicting and diagnosing algorithms or hardware behavior. Beyond Language Models Read the paper Language Models in Specialized Domains : Integrates music understanding and generation capabilities into LLMs, demonstrating LLMs’ potential in music composition. ChatMusician Read the paper : Aims to bridge LLMs’ gap in interpreting structured data, enhancing their ability to ground knowledge in tables, graphs, and databases. StructLM Read the paper : Focuses on responsibly creating Code LLMs, contributing to advancements in coding benchmarks, and emphasizing model openness. StarCoder2 and The Stack v2 Read the paper : Discusses video generation’s potential as a unified interface for diverse tasks, outlining challenges and future directions. Video as the New Language for Real-World Decision Making Read the paper Enhancing and Merging Language Model Capabilities : Proposes a method to fuse knowledge from multiple chat models, improving chat model performance through a novel merging technique. FUSECHAT Read the paper : Details a multilingual language model that showcases superior performance in coding tasks and multilingual capabilities. Nemotron-4 15B Technical Report Read the paper : Explores latent multi-hop reasoning in LLMs, revealing their inherent capabilities and limitations in complex reasoning tasks. Do Large Language Models Latently Perform Multi-Hop Reasoning? Read the paper Scaling and Efficiency in Model Training : Discusses a system for training LLMs on over 10,000 GPUs, tackling efficiency and stability challenges in large-scale model training. MegaScale Read the paper : Proposes a theory for optimizing LLM learning, aiming for reduced training steps and improved performance. Towards Optimal Learning of Language Models Read the paper : Introduces a model combining gated linear recurrences with local attention, offering an efficient alternative for language processing tasks. Griffin Read the paper : Investigates the effects of scaling on fine-tuning LLMs, providing insights into data, model, and method impacts on bilingual tasks. When scaling meets LLM finetuning Read the paper Improving Robustness and Diversity in AI : Generates diverse adversarial prompts to enhance LLM robustness, employing an open-ended search method for prompt discovery. Rainbow Teaming Read the paper : Proposes a deterministic sampling technique for code generation, improving sample diversity and model performance in compiler optimization. Priority Sampling of Large Language Models for Compilers Read the paper Generative Models and Interactive Environments : Trains a generative model to create interactive virtual worlds from various inputs, advancing generative AI and simulation capabilities. Genie r Read the pape Enjoyed This Story? I write a weekly analysis of the AI world in the Turing Post newsletter. Subscribe for free using the button below and be the first to read the latest stories. The goal of Turing Post is to equip you with comprehensive knowledge and historical insights, so you can make informed decisions about AI and ML. Join over 50,000 readers from the main AI labs, forward-thinking startups, and major universities 👇🏼

Explainable AI and Prompting a Black Box in the Era of Gen AI

Last Monday, a nurse suggested we try a wireless monitor to track my vitals and those of my unborn baby. the nurse told me. “We call this device “Monica, the monitor!” It’s either a dream to work with or a total nightmare,” On that day, “Monica” (actually the Novii Wireless Patch System) performed exceptionally well. I was able to move freely, without the encumbrance of wires, while giving birth to my daughter. This technology harnesses passive signal acquisition to differentiate between fetal and maternal heart signals and to detect uterine contractions. Data is wirelessly transmitted to a monitoring unit for real-time observation. This system enhances accuracy and reduces false alarms, offering so much-needed mobility during labor. I thought: that writing and theorizing about technologies is one thing, but experiencing their remarkable capabilities firsthand is quite another, especially when a device functions flawlessly. A question arose: What can foundation models add to wearables? Right after my experience with “Monica”, a recent paper from Google Research and MIT researchers caught my attention. Titled ‘ ,’ and authored by Kim et al., this paper delves into the application of LLMs in the health sector, focusing on interpreting data from wearable sensors for health prediction. Intriguingly, these models are fed data not from medical records or doctor’s notes, but from wearable devices like Fitbits, which track daily steps, heart rate, sleep patterns, and more — akin to ‘Monica.’ Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data The research evaluated eight cutting-edge LLMs: Med-Alpaca, PMC-Llama, Asclepius, ClinicalCamel, Flan-T5, Palmyra-Med, GPT-3.5, and GPT-4, across six public health datasets. They conducted experiments on thirteen health prediction tasks related to mental health, activity, metabolism, sleep, and cardiac assessments. The team experimented with various methods, including zero-shot and few-shot prompting (teaching the model with minimal or no examples), instructional fine-tuning (tailoring the model to specific tasks), and even some parameter-efficient fine-tuning for computational efficiency. Particularly fascinating is the effectiveness of context enhancement in prompts, which involves adding user context, health knowledge, and temporal information. This approach yielded up to a 23.8% improvement in performance. Healthcare is an exceedingly sensitive field, but the potential benefits of generative AI for humans are immense, especially with the power of foundation models. Health-LLM explores the future where wearables are not just passive trackers but proactive health guardians. Another recent groundbreaking paper in healthcare comes from Stanford and Stability AI researchers, titled . The most fascinating aspect of this paper is the development of CheXagent, an advanced foundation model specifically designed for interpreting chest X-rays. This model uniquely combines a clinical LLM, a specialized vision encoder, and a vision-language bridging network, demonstrating exceptional performance in interpreting complex medical images. Its ability to outperform existing models in accuracy and fairness evaluations marks a significant advancement in medical imaging AI technology. It can save so much time! And possibly lives. CheXagent: Towards a Foundation Model for Chest X-Ray Interpretation (The newborn girl — Reason Leeloo Joy — sends her regards. We took a week off last week but are now back on track, exploring the AI world to understand how she and her four brothers will live in it and navigate it.) News from The Usual Suspects © Sam Altman and OpenAI OpenAI two new embedding models (text-embedding-3-small and text-embedding-3-large) and updated versions of GPT-4 Turbo, GPT-3.5 Turbo, and a text moderation model. The new embedding models represent content as numerical sequences, enhancing machine learning tasks like clustering or retrieval. They are also more efficient and cost-effective. released Meanwhile, Sam Altman is with Middle Eastern backers, including wealthy investors and chip fabricators like TSMC, to launch a new chip venture. This move aims to meet OpenAI’s growing semiconductor needs and reduce dependence on Nvidia. The venture’s structure is unclear, and it might be a separate entity or a subsidiary of OpenAI. in discussions Blackstone steps in Another big player is investing heavily in the AI revolution. Blackstone is building a $25 billion network of power-intensive data centers across America. Following its $10 billion acquisition of QTS, a major data center operator, Blackstone is developing massive facilities to meet tech giants' growing digital and AI demands. These projects, consuming electricity equivalent to millions of homes, are reshaping communities and sparking debates over resource use and local benefits. Despite challenges, including strained power supplies and public backlash, Blackstone views this venture as one of its potentially best investments, illustrating the increasing significance and complexity of data infrastructure in the AI era. Elon Musk, xAI and Tesla Elon Musk has been making headlines recently, seeking a from global investors in the Middle East, Hong Kong, Japan, and Korea. If successful, xAI’s valuation could reach $20 billion, surpassing Anthropic’s $18.4 billion but falling behind OpenAI’s $100 billion. However, Musk’s recent unless he secures 25% control has stirred dissatisfaction among current investors and might affect talks with potential new backers. Meanwhile, in a “Dojo” supercomputer at its Buffalo, New York facility, underscoring the company’s commitment to advancing AI technology. $6 billion investment for xAI threat to remove Tesla AI projects Tesla is planning a $500 million investment Google and Hugging Face The recently announced aims to make AI more accessible. It focuses on shared initiatives in open science and source, leveraging both Hugging Face’s open models and Google Cloud’s technology. The goal is to facilitate the development of AI technologies for a wider range of users and applications. partnership between Hugging Face and Google Cloud Meanwhile, Google Bard has on HuggingFace’s Chatbot Arena Leaderboard, overtaking GPT-4 and now only behind GPT-4 Turbo in the community-driven LLM rankings. ascended to the second position The freshest research papers, categorized for your convenience Model Compression and Efficiency : A technique for efficiently compressing large language models by removing parameters while retaining performance SLICEGPT →read the paper : Focuses on developing high-performing, multi-language code generation models with an extensive parameter range DeepSeek-Coder →read the paper : Introduces an efficient pre-training method for T5 models, reducing computational requirements SPACTOR-T5 →read the paper : A framework for accelerating large language model inference using multiple decoding heads MEDUSA →read the paper LLM Capabilities and Evaluation : Evaluates MLLMs for generalizability, trustworthiness, and causality across multiple modalities From GPT-4 to Gemini and Beyond →read the paper : Develops a multilingual LLM supporting over 500 languages, enhancing language model accessibility MaLA-500 →read the paper : Introduces a method for zero-shot detection of text generated by large language models Spotting LLMs with Binoculars →read the paper Multimodal and Specialized Models : Examines the decoding mechanism in masked autoencoders for improved image processing Rethinking Patch Dependence for Masked Autoencoders →read the paper : A comprehensive survey on the advancements and capabilities of multimodal large language models MM-LLMs →read the paper : Establishes a benchmark for evaluating large multimodal models in the Chinese context CMMMU →read the paper : Enhances vision-language models with advanced spatial reasoning capabilities SpatialVLM →read the paper AI Training and Data Generation Techniques : Explores training neural networks for universal prediction strategies, approaching Solomonoff Induction Learning Universal Predictors →read the paper : A Python library for flexible and reproducible data preparation in generative NLP Unitxt →read the paper : A method for generating high-quality, content-grounded synthetic data using large language models GENIE →read the paper : Investigates a token-free language model that learns directly from raw bytes MambaByte →read the paper : Enhances language models with a task-agnostic scaffolding technique for better performance Meta-Prompting →read the paper : An approach for aligning large language models with human preferences in reinforcement learning WARM →read the paper Language Models and Role-Playing : Presents a compact model integrating enhanced vision vocabulary for efficient visual information encoding Small Language Model Meets with Reinforced Vision Vocabulary →read the paper : Develops a method for role-playing dialogues using large language models Large Language Models are Superpositions of All Characters →read the paper : Introduces a collection of multilingual large language models for conversational applications Orion-14B →read the paper In other newsletters Great dive into Apple’s “Update on apps distributed in the European Union” from Hardcore Software Fun read from about Model Merging “When what seems like pure LLM black magic is supported by the literature” Interconnects Is This The Year Apple Awakens in AI? Madrona investors’ opinion. describes his experience in Davos and World Economic Forum. It’s about AI but in Ng’s signature humanistic style. Andrew Ng

FOD#38: AI May Be Ushering Us Into a New Era of Health Monitoring

FOD#38: AI হয়তো আমাদেরকে স্বাস্থ্য পর্যবেক্ষণের একটি নতুন যুগে নিয়ে যাচ্ছে

FOD#38: KI könnte uns in eine neue Ära der Gesundheitsüberwachung einleiten

FOD#38: La IA puede estar llevándonos a una nueva era de vigilancia de la salud

FOD#38 : L'IA pourrait nous faire entrer dans une nouvelle ère de surveillance de la santé

एफओडी#38: एआई हमें स्वास्थ्य निगरानी के एक नए युग में प्रवेश करा सकता है

FOD#38: AI は私たちを健康モニタリングの新時代に導くかもしれない

FOD#38: AI가 우리를 건강 모니터링의 새로운 시대로 안내할 수 있음

FOD#38: A IA pode estar nos conduzindo a uma nova era de monitoramento da saúde

FOD № 38: ИИ может открыть нам новую эру мониторинга здоровья

FOD#38: Yapay Zeka Bizi Yeni Bir Sağlık İzleme Çağına Götürüyor Olabilir

FOD#38: AI có thể đưa chúng ta bước vào kỷ nguyên mới về theo dõi sức khỏe

FOD#38：人工智能可能将我们带入健康监测的新时代

Mobula API is a powerful tool that allows developers to access curated datasets for various crypto assets, such as market data, wallet transactions, historical balances, and swap quotes. Mobula API provides a unified and aggregated interface to query data from multiple blockchains and sources. In this article, you will learn how to use Mobula API to fetch and display market data for any assets using React, a popular front-end library for building user interfaces. Mobula API Overview Mobula API is a blockchain data provider that offers developers a wide range of features and functionalities. It provides curated datasets, market data with Octopus, wallet data, metadata with Metacore, and REST, GraphSQL & SQL interfaces to query them. Mobula API supports several chains, including: Ethereum, Avalanche, Polygon, Fantom, Harmony, xDai, Cronos, DFK Subnet, SmartBCH, Celo, Klaytn, Aurora, HECO, Arbitrum, And more. Developers can access vital data and perform key operations using Mobula API’s endpoints, designed to provide seamless integration and efficient access to blockchain features. Mobula API also offers easy-to-integrate tools to fetch market data, wallet transactions, holdings, etc. Retrieving Bitcoin Market Data Key Requirements for API Calls: API Key: Mandatory for production environments. Obtainable from your Mobula Dashboard. Not required for development mode. Method: Use GET to retrieve data. Base URL: https://api.mobula.io/api/1/ Endpoint: /market/query (for market data queries) Query parameters: Parameter Type Description Available Values filters string A comma-separated list of filters to apply to the data. liquidity, market_cap, volume, price, price_change_24h, price_change_1h, price_change_7d sortBy string A comma-separated list of fields to sort the data by. liquidity, market_cap, volume, price, price_change_24h, price_change_1h, price_change_7d sortOrder string The order in which to sort the data. asc (ascending), desc (descending) Example Usage: This example demonstrates how to retrieve market data for with and . mid-cap cryptocurrencies strong liquidity promising potential Parameter Breakdown: : filters : Selects assets with (inclusive). This ensures sufficient trading volume for easy entry and exit. liquidity:100:1000 liquidity between 100 and 1000 : Chooses assets with (inclusive). This range targets mid-cap projects with established value and room for growth. market_cap:10000:100000 market capitalization between 10,000 and 100,000 : Sorts the retrieved data by . This prioritizes larger and potentially more stable assets. sortBy=market_cap market capitalization in descending order : Specifies . This displays the largest cap currencies first, making it easier to identify prominent mid-cap contenders. sortOrder=desc descending order for sorting filters=liquidity:100:1000,market_cap:10000:100000
sortBy=market_cap
sortOrder=desc Response: {
    "name": "NFTDAO",
    "logo": "https://metacore.mobula.io/d0acfb425c161c8e3e5426c77ec36128ecfd688c2367a411cab9c440acb8b7a3.png",
    "symbol": "NAO",
    "liquidity": 769,
    "market_cap": 97230,
    "volume": 0,
    "price": 9.723276418969192e-10,
    "price_change_1h": 0,
    "price_change_24h": 0,
    "price_change_7d": 0,
    "contracts": [
      {
        "address": "0xd049206fb408a611e543791f2d8f102a8bc253dc",
        "blockchain": "Ethereum",
        "blockchainId": "1",
        "decimals": 18
      }
    ],
    "id": 4696,
    "rank": 4673
  } Explanation of the response: Field Description Value name Name of the cryptocurrency NFTDAO logo URL of the cryptocurrency’s logo 1 symbol Symbol of the cryptocurrency NAO liquidity Liquidity of the cryptocurrency 769 market_cap Market capitalization of the cryptocurrency 97230 volume Volume of the cryptocurrency 0 price Price of the cryptocurrency 9.723276418969192e-10 price_change_1h Percentage change in price over the last hour 0 price_change_24h Percentage change in price over the last 24 hours 0 price_change_7d Percentage change in price over the last 7 days 0 contracts List of contracts associated with the cryptocurrency [{“address”: “0xd049206fb408a611e543791f2d8f102a8bc253dc”, “blockchain”: “Ethereum”, “blockchainId”: “1”, “decimals”: 18}] id Unique identifier for the cryptocurrency 4696 rank Rank of the cryptocurrency based on market capitalization Exploring Block Data with Mobula Queries To build a React project that interacts with the Mobula API and fetches data, follow these steps: Step 1: Set Up Your React Project First, ensure you have Node.js and npm installed on your machine. Then, create a new React project using Create React App: npm create vite@latest mobula-market-app -- --template react
cd mobula-market-app Step 2: Install Dependencies Install Axios, a popular HTTP client for making API requests, to handle HTTP requests in your React project: npm install axios Step 3: Create a Component for API Interaction Create a new file named inside the directory to contain the logic for fetching market data: MarketData.js src // src/MarketData.js
import React, { useState, useEffect } from "react";
import axios from "axios";
import { Line } from "react-chartjs-2";

const MarketData = () => {
  const [marketData, setMarketData] = useState(null);
  const [selectedCrypto, setSelectedCrypto] = useState(null);
  const [filters, setFilters] = useState("");
  const [sortBy, setSortBy] = useState("");
  const [sortOrder, setSortOrder] = useState("");
  const [showModal, setShowModal] = useState(false);

  const fetchData = async () => {
    try {
      const response = await axios.get(
        `https://api.mobula.io/api/1/market/query?filters=${filters}&sortBy=${sortBy}&sortOrder=${sortOrder}`
      );
      setMarketData(response.data);
    } catch (error) {
      console.error("Error fetching data:", error);
    }
  };

  useEffect(() => {
    fetchData();
  }, [filters, sortBy, sortOrder]);

  const handleCardClick = (item) => {
    setSelectedCrypto(item);
    setShowModal(true); // Ensure showModal is set to true when a card is clicked
  };

  const handleCloseModal = () => {
    setShowModal(false);
  };

  const renderChart = () => {
    if (!selectedCrypto) return null;

    const data = {
      labels: ["1h", "24h", "7d"],
      datasets: [
        {
          label: "Price Change",
          data: [
            selectedCrypto.price_change_1h,
            selectedCrypto.price_change_24h,
            selectedCrypto.price_change_7d,
          ],
          fill: false,
          borderColor: "rgb(75, 192, 192)",
          tension: 0.1,
        },
      ],
    };

    return (
      <div
        style={{
          position: "fixed",
          top: "50%",
          left: "50%",
          transform: "translate(-50%, -50%)",
          backgroundColor: "#fff",
          padding: "20px",
          zIndex: 9999,
        }}
      >
        <button
          onClick={handleCloseModal}
          style={{ position: "absolute", top: "10px", right: "10px" }}
        >
          Close
        </button>
        <h3>Price Change Chart for {selectedCrypto.name}</h3>
        <Line data={data} />
      </div>
    );
  };

  const renderCryptoCards = (cryptoData) => {
    return cryptoData.map((item, index) => (
      <div
        key={item.id}
        style={{
          border: "1px solid #ccc",
          borderRadius: "5px",
          padding: "10px",
          marginBottom: "20px",
          width: "calc(33.33% - 20px)",
          marginRight: "20px",
          display: "inline-block",
        }}
      >
        <img
          src={item.logo}
          alt={`${item.name} logo`}
          style={{ width: "50px", height: "50px", marginRight: "10px" }}
        />
        <div>
          <h3>{item.name}</h3>
          <p>
            <strong>Symbol:</strong> {item.symbol}
          </p>
          <p>
            <strong>Liquidity:</strong> {item.liquidity}
          </p>
          <p>
            <strong>Market Cap:</strong> {item.market_cap}
          </p>
          <button onClick={() => handleCardClick(item)}>View Graph</button>
        </div>
      </div>
    ));
  };

  const renderBlockchainTabs = () => {
    if (!marketData) return null;

    const categorizedData = {};

    marketData.forEach((item) => {
      const blockchain = item.contracts[0].blockchain;
      if (!categorizedData[blockchain]) {
        categorizedData[blockchain] = [];
      }
      categorizedData[blockchain].push(item);
    });

    return Object.entries(categorizedData).map(([blockchain, data]) => (
      <div key={blockchain}>
        <h2>{blockchain}</h2>
        <div style={{ display: "flex", flexWrap: "wrap" }}>
          {renderCryptoCards(data)}
        </div>
      </div>
    ));
  };

  const handleSubmit = (e) => {
    e.preventDefault();
    fetchData();
  };

  return (
    <div>
      <form onSubmit={handleSubmit}>
        <label>
          Filters:
          <input
            type="text"
            value={filters}
            onChange={(e) => setFilters(e.target.value)}
          />
        </label>
        <label>
          Sort By:
          <select value={sortBy} onChange={(e) => setSortBy(e.target.value)}>
            <option value="">Select</option>
            <option value="liquidity">Liquidity</option>
            <option value="market_cap">Market Cap</option>
          </select>
        </label>
        <label>
          Sort Order:
          <select
            value={sortOrder}
            onChange={(e) => setSortOrder(e.target.value)}
          >
            <option value="">Select</option>
            <option value="asc">Ascending</option>
            <option value="desc">Descending</option>
          </select>
        </label>
        <button type="submit">Fetch Data</button>
      </form>
      {renderBlockchainTabs()}
      {showModal && renderChart()}
    </div>
  );
};

export default MarketData; Step 4: Integrate the Component in Your App Modify the file to include the component: App.js MarketData // src/App.js
import React from 'react';
import './App.css';
import MarketData from './MarketData';

function App() {
  return (
    <div className="App">
      <header className="App-header">
        <h1>Mobula Market Data</h1>
      </header>
      <main>
        <MarketData />
      </main>
    </div>
  );
}

export default App; Folder structure: Step 5: Run Your React App Start your React development server to see the app in action: npm start

Using Mobula to Fetch and Display Market Data For Assets

214 Stories To Learn About Computer Vision

Stories, News & Mentions about X2AI

X2ai’s Stories on HackerNoon