Kasisto

kasisto

The emergence of (AI) has ushered in an era of unprecedented change across numerous industries, and the cloud contact center space is no exception. With the promise of enhancing efficiency, improving customer interactions, and providing valuable insights, AI is rapidly reshaping the landscape of cloud contact centers, transforming them into intelligent hubs of communication. Artificial Intelligence Today, AI's footprint is increasingly evident, as it allows cloud contact centers to manage high call volumes, automate routine tasks, and deliver real-time analytics. , the market for artificial intelligence (AI) in call centers is projected to grow significantly, expanding from approximately $800 million in 2019 to around $2.8 billion by 2024. According to a report by Statista This article delves into the fascinating journey of AI’s integration into cloud contact centers, examining the past, present, and future implications. The Past: Traditional Contact Centers In the past, traditional contact centers primarily operated on voice-based telecommunication systems and physical infrastructures, requiring substantial human resources to manage and maintain customer interactions. Following are the key aspects of the past traditional contact centers: Structure of Traditional Contact Centers Traditional contact centers, the predecessors of cloud contact centers, were primarily composed of human operators who handled customer queries, and complaints, and provided support services. They served as a bridge between the company and customers. The operations were conducted through telephone calls and later, emails, necessitating a vast network of operators and infrastructure. Limitations of Traditional Contact Centers While these centers played a crucial role in customer service, they had several limitations. Inefficiencies in service were prevalent, with customers often experiencing long wait times and sometimes, inconsistent service due to human error. The scalability of these centers was also a concern, with the need to physically expand facilities to accommodate growth. Additionally, handling vast amounts of data and deriving meaningful insights was challenging due to the lack of sophisticated data analysis tools. The Shift: AI and Cloud Contact Centers Cloud contact centers emerged as a solution to these challenges. As the name suggests, these centers operate on cloud technology, reducing the need for extensive physical infrastructure. They offer numerous benefits including cost-effectiveness, scalability, flexibility, and business continuity, transforming the customer service landscape. AI's Inception into Contact Centers Alongside the emergence of cloud technology, AI began to permeate the contact center industry. Initial applications included voice recognition systems and automated responses, aiming to reduce the workload on human operators and provide quicker, more efficient service. The Present: Integration of AI in Cloud Contact Centers The integration of AI into cloud contact centers brings exciting advancements, including AI chatbots, AI voice assistants, and predictive analytics. These innovations enhance efficiency, scalability, and customer-centricity, revolutionizing contact center operations. Role of AI in Modern Cloud Contact Centers AI plays a pivotal role in modern cloud contact centers, revolutionizing customer service through AI chatbots, AI voice assistants, and predictive analytics. AI Chatbots Chatbots are one of the most visible applications of AI in cloud contact centers. These automated virtual assistants handle customer queries via text conversations, providing instant responses and operating 24/7. They can manage multiple queries simultaneously, offering a level of scalability far beyond human operators. highlights the benefits of AI chatbots in contact centers, citing that by 2027, chatbots will become the primary customer service channel for roughly a quarter of organizations. The use of chatbots can support better customer experiences, especially for simple, repetitive tasks that take up most of a contact center agent’s time. Forbes AI Voice Assistants AI voice assistants, such as have also found their place in modern contact centers. Using advanced voice recognition and natural language processing capabilities, they offer a more interactive and personalized customer service experience. Amazon's Alexa or Google Assistant, Predictive Analytics Another critical application of AI in contact centers is predictive analytics. By analyzing past data, AI systems can predict future customer behavior, allowing operators to offer personalized services and proactive solutions. This enhances both customer satisfaction and operational efficiency. AI at Work: Real-world Applications and Success Stories AI's impact on cloud contact centers is not a speculative discussion; it's already creating success stories. Here are two instances: has leveraged Watson, their AI technology, within their contact center operations, providing personalized customer service through advanced NLP and machine learning. The results have been a dramatic improvement in customer satisfaction levels and operational efficiency. IBM has unveiled its Contact Center AI solution that uses AI to provide real-time assistance to human agents, leading to more efficient and higher-quality customer interactions. Google has leveraged AI to enhance its customer service by introducing Mastercard KAI, an AI-powered chatbot developed in partnership with Kasisto, designed to offer banking and merchant services, which was initially rolled out on Facebook Messenger. The company also improved its product content to boost Customer Satisfaction Score (CSAT), implementing solutions from Zoomin and IXIASOFT to streamline content operations, improve the user search experience, and deliver personalized content at scale. Mastercard The Future: Trends and Predictions for AI in Cloud Contact Centers With advancements in AI and an increasing emphasis on superior customer service, the future of AI in cloud contact centers is full of exciting possibilities. : As AI technologies mature, expect the level of personalization in customer service to reach unprecedented heights. AI will leverage extensive customer data to tailor experiences to individual preferences, past interactions, and behaviors. Hyper-personalized Customer Service : AI's predictive capabilities will be harnessed to anticipate customer needs and behavior patterns, leading to more proactive service. This trend could see a reduction in incoming calls and higher customer satisfaction as businesses can resolve issues before they escalate. Predictive Customer Service As AR technology becomes more refined, it could be combined with AI to offer a more interactive customer service experience. Visual assistance powered by AR could guide customers through solving their issues, reducing the need for agent intervention. : Augmented Reality (AR) Support Embracing AI: A Competitive Imperative As the role of AI expands in cloud contact centers, businesses have to adapt to keep pace with evolving customer expectations. Leading cloud contact center providers, such as and , harness the power of AI to drive growth, demonstrating the significance of AI integration in this industry. The integration of AI is not just about cost-saving or streamlining operations; it's about leveraging it as a strategic asset that enhances customer relationships, drives innovation, and creates a competitive advantage. Teloz RingCentral Powering the Next Chapter in Cloud Contact Center Evolution AI's transformative role in cloud contact centers is here to stay, and it's only set to become more prominent in the coming years. By providing better customer service, managing high call volumes, and enabling real-time analytics, AI is helping businesses to redefine the customer contact center experience. As we move toward the future, AI will continue to catalyze significant changes in the landscape, offering an exciting world of possibilities for customer interaction. This revolution is just the beginning, and the most innovative businesses will be those that harness the power of AI to transform the customer experience in their cloud contact centers.

The Evolving Role of AI in Cloud Contact Centers

El papel en evolución de la IA en los centros de contacto en la nube

Le rôle évolutif de l'IA dans les centres de contact cloud

क्लाउड संपर्क केंद्रों में एआई की उभरती भूमिका

クラウド コンタクト センターにおける AI の役割の進化

O papel em evolução da IA nos contact centers na nuvem

Vai trò đang phát triển của AI trong Trung tâm liên hệ trên đám mây

人工智能在云联络中心中不断演变的作用

This is today's reality: Artificial intelligence has already made a lot of buzz in the mobile app development industry. More cheap and available screens, easy real-time access to the data robust analysis tools have become even more powerful - all of this is already a normal part of our daily routine in society. We're approaching the stage of smart everything The implementation of Artificial Intelligence is highly varied, depending on specific industries. Chatbots, spam filters, cloud APIs, natural language processing, face recognition robotics and more are expected to be used for many goals. For example, installing chatbots into retail screens and POSs, inserting built-in intelligence to popular business applications, and integrating with an environment-aware smart detector. And, what is most important, AI will become such a normal thing in everyday life that it's already used in small shops and many industrial factories. AI will become more affordable Apple is taking a shot at a processor planned exclusively to perform AI-related errands, and it will, in the long run, be remembered for huge numbers of its smart gadgets, including iPhones and iPads. Curiously, this is one case where Apple isn't driving the way - partly because it's playing catch up for the lost time. Qualcomm, as of late, discharged the Snapdragon 835 portable processor, which has a module for AI and taking care of man-made brainpower assignments. Snapdragon processors are utilized in numerous Android telephones. Google revealed an AI-related processor back to the Tensor Processor Unit in 2016, in spite of the fact that, until further notice, the chip is utilized in Google server farms for picture acknowledgment and to convey quicker, better query items. Don't be astounded if Google builds up an AI-centered portable processor too. Simultaneously, the cost of sensors has come down, making them affordable for the cheapest vehicles and devices, yet it can be used for anything – from modern machines to shoes, to drink bottles. Increasingly reasonable sensors, implanted in everything from apparel to hardware, will add data to a regularly developing system of associated gadgets. For instance, Wearable X reported $299 yoga pants that will use implanted sensors and a versatile application to help yogis of all capacities to improve their flexibility, etc.. Indeed, even the most costly classes of sensors are seeing costs fall. In April, Velodyne reported another, minimal effort strong state LiDAR framework called Velarray, which could make self-driving autos a modern reality. The sensor is equipped for perceiving objects with low reflectivity up to 200 meters away, flaunts improved vertical and flat fields of view than forerunners, and expects to sell for only two or three hundred bucks when delivered in mass volumes. We'll likely see sensors transmitting information on everything from street conditions, to office use and modern hardware. Undertakings, which are utilizing more mobile phones and wearables than any time in recent history, should transmit, gather and investigate this information to settle on always making smart choices – and they'll be needing to do it progressively. Keen hardware representatives with smartphones on the shop should be fixed, sensors cautioning field administration workers with smartphones that they have to make a quick help call.  This is only the tip of the iceberg. This new equipment is clear proof that AI is genuinely turning into an ordinary reality. Be that as it may, while the equipment improvements are basic, the other piece of the story is the product – or the applications. When we gather the information from every one of these sensors, AI will help examine and instruct representatives on the next steps, yet the venture work environment has changed. Representatives have moved away from work area PCs and paper-and-clipboards and are accomplishing increasingly more business on their smartphones. Conventional business application designers must think of AI as far as versatile applications. What separates mobile apps out from web applications on PCs is their capacity to know about their location and the external world utilizing these sensors. Combining AI with these abilities makes applications significantly more valuable. Here are several cases of AI and machine learning contributions from the sensors that make proposals on what you ought to do alongside with the intensity of a portable application: AI implementation across the industries Buoy Health is an AI-based symptom and medicine controller that applies computer algorithms to identify and cure sicknesses. The principle: a chatbot receives a person’s symptoms and information from their medical companies.  The AI then leads that person to the right protection in accordance with its analysis. Harvard Medical School already uses Buoy’s AI to diagnose and immediately and more smoothly handle many cases. AI implementation in Healthcare creates deep learning medical applications to modernize radiology examinations. The AI platform interprets disorganized therapeutic info (radiology pictures, blood analyses, EKGs, patient therapeutic records) to provide doctors with more useful penetration into a patient’s interactive requirements. Enlitic Komatsu offers a solution, SmartConstruction, that associates work data concerning gear and individuals. SmartConstruction enables clients to review their workplace data by means of automaton or 3D scanning to deliver a 3D model of current conditions, bringing the information into KomConnect to consequently recreate the making of development plans, and afterward transmitting the information into mechanized, driverless, earth-moving vehicles to computerize exhuming. Komatsu's shrewd machine control can wipe out human mistakes and forestall security perils. Simultaneously, Komatsu's versatile application enables clients to see machine data from the Komtrax armada observing framework on their Apple or Android cell phones or tablet gadgets. AI implementation in Development Kasisto has built an AI platform, KAI, which allows banks to create their own personal chatbots and virtual assistants. It’s entrenched in AI logic and human-language recognition and creation, which would imply that it can deal with complex issues about finance and banking service. AI implementation in Banking Here is another example of using AI in financial applications: Simudyne is a tech vendor that applies AI and machine learning to reach thousands of business situations. Simudyne's program enables financial companies to manage resistance testing examinations and manage the risks for market contamination on huge ranges. The CEO of the corporation announced that the simulation supports investment managers to avoid risks and improve critical debt provisioning for the purpose of making better-informed judgments. How to apply AI in current mobile applications Even a few years ago, many revolutionary and innovative technologies needed years to be implemented to business, however, today you can begin easily and reach your goals much more quickly. Business developers all across the globe are already supporting businesses to enhance independence and modernize enterprise processes with restricted resources, by using AI with modern UI design in current applications. How can developers begin to use artificial intelligence for their applications to enhance UX and reduce the chances of failure: Traditionally, signals from mobile apps are set on a timer, but AI provides alerts or warnings set on customers' circumstances or interactive actions. Real-time and Immediate Alerts Applications can make smart, personalized notes relied on various operations and actions that are obtained and then interpreted. Custom Notification Apps can read the shopping habits and routines of customers and are able to provide owners with valuable information to aid them in their business operations. Self-Reading Applications that use AI can do even more than “enough” to create a comfortable and safe environment. For example, such an app can turn off an air conditioner or electric outlet even when a person is not in the house. Smart Actions When AI is integrated into a search engine it is able to make more habitual and less painful processes for mobile users to fasten a search, suggest searching options and grammar improvements, and amplify voice search.. To sum up, applying Artificial Intelligence (AI) to a mobile app can constantly help to streamline the searching process for mobile users. Streamline the searching process For the largest app development corporations, fingertips and face recognition are the most common AI technologies. Artificial intelligence reads eye position, nose size, a form of the lips, etc. Those people who are using the latest iPhones are already familiar with facial identification technology. Google goes further with FaceNet. On the broadly applied Labeled Faces in the Wild (LFW) dataset, they reached a new high certainty of 99.63%. Using an AI and a smart algorithm has managed to associate a face to its owner with excellent accuracy. Face recognition and authentication As reported by Gartner in the , the amount of organizations realizing AI technologies has increased 2.7 times from 2015. 2019 CIO Survey Artificial intelligence (AI) is the most discussed technology and one-ups the data and analytics as a ground-breaking technology. In terms of realization, when told about the company’s projections about coming digital technologies and tendencies, 37% answered that they already extended AI. Actually, AI appears in second, behind cybersecurity (88 %). The Future of Artificial Intelligence in Mobile App Development Based on the Statista report, by the year 2025, the growth of the artificial intelligence (AI) software market worldwide will reach $89 billion. However, the embedding of AI in the mobile application is in the initial phases. There is a lot that can still be changed and modified. Your complete UX can be formed by applying AI and then handled in various mobile apps. We can see a remarkable increase in AI use in 2020 already. AI has already started to integrate into 5G smartphones which means that the customers can get more privileges of ML and AI to gather, maintain, and handle real-time information. With the purpose of getting the new impressive functions in a mobile app, you should think about choosing mobile developers for your company. Here at KitRUM , we're always happy to hear from you and help you to get access to the top 5% of most talented application developers. Previously published at https://kitrum.com/blog/artificial-intelligence-in-mobile-app-development-upcoming-trends-for-2020/

Artificial Intelligence in Mobile App Development in 2020

If you missed the first AI industry analysis on the Insurance vertical , check that out here . I. Financial Innovation: Lots of Talk, Little Action? TThe financial sector is historically one of the most resistant to change you might think of. It is then inevitable that big banks from one hand and startups from the other hand are creating a huge break in the financial industry and I believe this is happening not because of the use of a specific technology but rather because of their intrinsic cultural differences, diverse structural rigidity, and alternative cost-effective business models. In other words, banks do not innovate either because they are too big to quickly adapt and follow external incentives or because they don’t know how (and want to) truly change. This is not simply true in the industry but also in academia, where until the mid-nineties there were no relevant contributions to financial innovation at all (Frame and White, 2002). In fact, in few survey articles (Cohen and Levin, 1989; Cohen, 1995) with more than 600 different articles and books quoted, . none of them was related to financial innovation subjects Of course, things changed over the last five years, but my opinion is that was really out of necessity rather than a voluntary push-approach from the banking sector. Financial innovation is, therefore, something which seems to be usually rather than , and often more characterized by a rather than a one (although this might be a controversial opinion, I guess). Given the new technological paradigm (which is tightening the inner strong causal relationship between innovation and growth) it seems natural to wonder whether a better innovation model can be therefore imported by a different (and more successful) sector. imported internally generated product-innovation process I found that there is a very specific and interesting case of a sector which had to ‘ ’ rather than ‘ ’: the (Baker, 2003; Gans and Stern, 2004; Fuchs and Krauss, 2003; Lichtenthaler, 2008). innovate-to-survive innovate-to-grow biopharma industry Innovation drivers map. The biopharma companies feel more the urgency to innovate and are also more committed to that . The graph I built plots 25 major banks (blue) and 25 major pharmaceutical (red) companies based on their Innovation Impulse and Commitment. The Impulse variable has been built using the number of patents a company filed (a proxy for the external pressure to innovate) and the number of recorded shareholders (a proxy for the internal pressure to innovate). The Commitment shows instead the R&D intensity (net sales) while the size of the bubbles the net income for each company. The data points were obtained by Medtrack, Osiris, and Zephyr in 2014. II. Innovation Transfer: the Biopharma Industry The biopharma industry is not a single sector but actually two different ones: , populated by smaller companies that drive the research and exploration phase, and the , big giants that through the last century became huge go-to-market and sales enterprises. the biotech space pharmaceutical companies Hence, there is from one hand and from the other…Is it something that we have already seen somewhere, didn’t we? pure (risky) innovation pure commercialization skills The biopharma industry and the financial sector suffer from a strong polarized innovation What characterizes the industry is that the risky activity lies in the initial development process rather than in the market phase. The problem is not to match customer demand or find a market for your product, but it is actually developing the molecule in the first place. The probability of success is extremely low and the timeline very long (10–15 years) and the 20-years patents give you only a temporary advantage. More importantly, it looks that only 3 out of 10 of the drugs produced are able to repay the development costs (Meyer, 2002) and that most of the companies operate at loss while the top 3% companies alone generate almost 80% of the entire industry profits (Li and Halal, 2002). A tough business, isn’t it? The biopharma industry is then no longer simply a but also a . Innovation is not ancillary but it is the quintessential driver to survive. And this is also why they had to identify a range of different methods to foster their growth-by-innovation: R&D, competitive collaboration schemes, venture funding, co-venture creation, built-to-buy deals, limited partnership agreements, etc. human-intensive business capital-demanding one It should be clear by now where I am heading to: as the biopharma sector and it is and make it profitable. the financial industry doesn’t strongly feel the need to innovate not experimenting and pushing to create new models that might spread their innovation risk The financial industry doesn’t strongly feel the need to innovate III. Introducing AI, Your Personal Financial Disruptor Image Credit: Haru1/Shutterstock By now you might object “All good man, but FS and biopharma are still sooo different, so why should I import innovation models from a sector which is completely different from mine?”. Well, that’s the catch: I don’t think they are. And the reason why they are becoming a lot more similar is precisely Artificial Intelligence. and has a which are somehow similar to the biopharmaceutical one: it requires a long time to be created, implemented and correctly deployed (with respect to the financial industry standards, of course); it is highly technical and requires highly specialized talents; it is highly uncertain, because you need to experiment a lot before finding something that works; it is expensive, both in terms of time as well as monetary investments (talents, hardware, and data are really expensive); it is risky and the risk lies in the initial development phase, with a very high-payout but a high likelihood to fail as well. AI is creating a strong pressure to innovate for the financial sector development cycle and characteristics AI is creating a strong pressure to innovate for the financial sector But AI is also introducing a completely new speed and degree of trust in the financial industry, which lowers the tolerable mistakes at the same level of the biopharma sector. If your algorithms point out to the wrong product to sell or the wrong book to be recommended, it is not a big deal. If your system misinterprets some signals in the market or while developing a drug though, you end up losing millions in seconds or even losing human lives. It is then not only stretching out issues that intrinsically belong to the financial sector such as or , but it is also bringing new problems such as to the picture (specifically in consumer applications). regulation accountability or the biased data lack of transparency And last but not least, AI is making the question mark on the matter bigger than even in FS, the same as it was in the biopharma industry back in the nineties and that culminated in the current biotech-pharmaceutical dichotomy (if you are wondering anyway, this choice is all focused around on your , , and with respect to your competitors — do you have enough data to train an ANI? Can your team/project scale? Is the ANI unique or something your peers are doing or need to do as well?). “build vs buy” data capacity team and project scalability uniqueness of the project AI is revolutionizing a centuries-old industry innovation flow from the ground This is why I believe AI in financial services to be extremely important — not much for the specific innovation or product it is introducing but rather because . it is revolutionizing a centuries-old industry innovation flow from the ground IV. Segmentation of AI in Fintech Image Credit: TZIDO/Shutterstock Artificial Intelligence is using structured and unstructured data in financial services to improve the customer experience and engagement, to detect outliers and anomalies, to increase revenues, reduce costs, find predictability in patterns and increase forecasts reliability…but it is not so in any other industry? We all know this story, right? So what is really peculiar about AI in financial services? First of all, FS is an industry full of data. You might expect this data to be concentrated in big financial institutions’ hands, but most of them are actually and thanks to the new to smaller players as well. AI can then be easily developed and applied because the barriers to entry are lower with respect to other sectors. public EU payment directive (PSD2) larger datasets are available Second, many of the underlying processes can be relatively while many others can be improved by either brute force computation or speed. And historically is one of the sectors that needed this type of innovation the most, is incredibly competitive and is always looking for some new source of ROI. easier to be automatized Bottom line: the marginal impact of AI is greater than in other sectors. Third, the makes the field really fertile for AI. AI needs (a lot of) innovative data and above all feedback to improve, and millennials are not only happy to use AI as well as providing feedback, but apparently even less concerned about privacy and giving away their data. transfer of wealth across different generations There are also, of course, a series of in financial sector that limit a smooth and rapid implementation: legacy systems that do not talk to each other; data silos; poor data quality control; lack of expertise; lack of management vision; lack of cultural mindset to adopt this technology. specific challenges for AI So what is missing now is only having an overview of the AI fintech landscape. There are also plenty of maps and classification of AI fintech startups out there (probably the best ones are the one provided by , in particular and ), so I am not introducing anything new here but rather simply giving you my personal framework: CB Insights this this : this category is about making the end-client life better and easier and it includes ; ; automated financial advisors and planners that assist the users in making financial decisions ( check ); smart wallets that coach users differently based on their habits and needs. Financial Wellness personalized financial services credit scoring robo-advisor, virtual assistants, and chatbots — CB Insights map if you like Examples include [robo-advisors and conversational interfaces] Kasisto ; Trim ; Penny ; Cleo ; Acorns ; Fingenius ; Wealthfront ; SigFig ; Betterment ; LearnVest ; Jemstep ; [credit scoring] Aire ; TypeScore ; CreditVidya ; ZestFinance ; Applied Data Finance ; Wecash ; : I think that, given the importance of this instrument, it deserves a separate category regardless of the specific application is being used for (which may be payments, compliance, trading, etc.). Blockchain Examples include: Euklid ; Paxos ; Ripple ; Digital Asset ; : this can be divided into (payment security and physical identification — ) and (looking for fraudulent and abnormal financial behavior — ). Financial Security identification biometrics and KYC detection AML and fraud detection Examples include, respectively: EyeVerify ; Bionym ; FaceFirst ; Onfido ; and Feedzai ; Kount , APEX Analytics ; this category includes payments, peer-to-peer lending, and debt collection. Money Transfer: Examples include: TrueAccord ; LendUp ; Kabbage ; LendingClub ; : this is a big section, and I tend to divide it into five main subsections: Capital Markets (either algotrading or trading/exchange platforms). i) Trading Examples include: Euclidean ; Quantestein ; Renaissance Technologies , Walnut Algorithms ; EmmaAI; Aidyia ; Binatix ; Kimerick Technologies ; Pit.ai ; Sentient Technologies ; Tickermachine ; Walnut Algorithm ; Clone Algo ; Algoriz ; Alpaca ; Portfolio123 ; Sigopt ; (either crowdsource funds or home-trading). ii) Do-It-Yourself Funds Examples include: Sentifi ; Numerai ; Quantopian ; Quantiacs ; QuantConnect ; Inovance ; (information extraction or insights generation). iii) Markets Intelligence Examples include: Indico Data Solutions ; Acuity Trading ; Lucena Research ; Dataminr ; Alphasense ; Kensho Technologies ; Aylien ; I Know First ; Alpha Modus ; ArtQuant ; (most of the alternative data applications are in capital markets rather than broader financial sector so it makes sense to put it here). iv) Alternative Data Examples include: Cape Analytics ; Metabiota ; Eagle Alpha ; v) (this section is more a residual subcategory because most of the time startups in this group fall within other groups as well). Examples include: Risk Management Ablemarkets ; Financial Network Analysis . Conclusions I am arguing since the beginning of the article that AI is making financial services and biopharma much more alike, and that the FS industry might learn something from how the other industry innovates. The reality is that the financial industry has also very specific traits and challenges it needs to overcome. The biggest difference I currently see in that is the effect AI is having on the physical products market: while in almost any sector AI is used with the final goal of creating or improving new products (and this is true also for drug development, for example) in the financial ecosystem is having exactly the opposite effect. . Its final goal will be to create the (frictionless) bank of the future: no branches, no credit cards, no frauds, no menial reporting activities. A with modular components that increases our financial literacy and has no physical products or spaces. AI is making the industry more digitalized than ever before bank-as-a-platform It would definitely be a great world to live in. Can’t wait for it. Reference Baker, A. (2003). “Biotechnology’s Growth-Innovation Paradox and the New Model for Success”. 9 (4): 286–88. Journal of Commercial Biotechnology Cohen, W. (1995). “Empirical Studies of Innovative Activity”, in , edited by Paul Stoneman. Cambridge, Mass.: Blackwell. Ch. 6, 182–264. Handbook of the Economics of Innovation and Technological Change Cohen, W., Levin, R. (1989). “Empirical Studies of Innovation and Market Structure”, in , Vol. 2, edited by Richard Schmalensee and Robert Willig. Amsterdam: North-Holland. Ch. 18, 1059–1107. Handbook of Industrial Organization Frame, W. S., White, L. J. (2002). “Empirical studies of financial innovation: lots of talk, little action?”. Working Paper, Federal Reserve Bank of Atlanta, N. 2002–12. Fuchs, G., Krauss, G. (2003). “Biotechnology in Comparative Perspective”. In: . G. Fuchs (ed.). New York: Routledge, 1–13. Biotechnology in Comparative Perspective Gans, J., Stern, S. (2003). “Managing Ideas: Commercialization Strategies for Biotechnology”. 01/03: 1–24. Intellectual Property Research Institute of Australia Working Paper Li, J., Halal, W. E. (2002). “Reinventing the biotech manager”. , 20 Suppl (6): 61–3. Nature Biotechnology Lichtenthaler, U. (2008). “Open Innovation in Practice: An Analysis of Strategic Approaches to Technology Transactions”. , 55 (1): 148–157. IEEE Transactions on Engineering Management Meyer, F. J. (2002). “Business Models That Biotech Companies Employ”. November 25, 2002 Enterprise Development KFBS Biotech Speakers Series — — Follow me on Medium Look at my other articles on AI and Machine Learning: Unsupervised Investments (I): A Guide to AI Investors Unsupervised Investments (II): A Guide to AI Accelerators and Incubators Who is best positioned to invest in Artificial Intelligence?

Artificial Intelligence Verticals (II): Fintech

Unsupervised Investments (I): A Guide to AI Investors

DAOs are eating the world. Already, there are hundreds of DAOs formed around several causes, from and providing to media production. enabling social interaction decentralized finance services While the infamous DAO hack of 2016 dampened enthusiasm for blockchain governance, the growth of several projects shows DAOs can become a new force in traditional economies. But before that can happen, DAOs must solve a teething problem: governance. Every organization, even decentralized ones, must coordinate to decide on critical issues. The success or failure of any DAO depends on how well it can achieve governance objectives in a secure, transparent, and fair manner. Why Does Governance Matter For DAOs? In traditional organizations, the Board of Directors or a smaller group of C-Suite executives are usually responsible for decision-making. While centralized control promotes efficiency, years of management problems have revealed flaws inherent in traditional governance. One such flaw is the so-called "principal-agent problem." In economics, the principal-agent problem occurs when the interests of individuals (principals) and representatives deciding on their behalf (agents) are misaligned. For example, directors may chase short-term profits, which benefit them, at the expense of long-term growth that benefits investors. Decentralized autonomous organizations were formed to democratize decision-making. By distributing power across the organization, DAOs hope to align the interests of members and ensure that no individual or group can pursue selfish objectives. DAOs primarily achieve this aim by using tokens to facilitate governance. Buying tokens signals a commitment to the DAO, and by extension, fitness to participate in governance. Holding a DAO token imbues the holder with significant influence on the organization's direction. For example, DAO members can call for changes by submitting a proposal. In other words, a token functions like a share in a traditional company. Armed with tokens, DAO members can express their opinion on several issues requiring consensus. Usually, DAO voting sessions are executed on-chain, which provides a higher level of transparency and security than regular balloting systems. Examples of governance tokens. [ ] Image source DAO governance can serve some or all of the following purposes: : DAOs often raise initial capital via token sales, with the remaining funds deposited in a treasury. However, as the DAO grows its user base and cash flow, it needs to manage the allocation of funds for projects. Asset Management : Even a DAO needs a group of people to lead certain areas. However, the difference is token holders have full control of the election and ouster of officials. For example, the Ethereum Naming Service (ENS) DAO recently asked members to for the removal of its director, Brantly Millegan. HR Processes vote on a proposal : Governance tokens can be used to vote on proposals concerning the operation of a network. For example, members of the TerraDAO to burn an agreed number of Luna tokens and mint more TerraUSD stablecoins. Protocol Decisions voted Governance is a critical part of DAO infrastructure. Even though these organizations have fluid structures, a decision-making framework is important. Otherwise, it may be impossible to reach a consensus on matters critical to the DAO's growth. Challenges With Traditional DAO Voting Protocols Most early DAOs adopted a token-based system of governance ("one token, one vote"). In this system, higher influence on DAO governance required a higher level of financial commitment. This arrangement ensured individuals with a proven commitment to the health of the DAO had greater input in running it. Token-holder voting systems provide a simple, safe, and transparent way for DAOs to make decisions. Once a proposal is up for discussion, members can cast their "votes" on-chain and verify the results. There are no hidden election processes; everything is right there on the blockchain for everyone to see. But, as innovative as they are, 1T1V systems may pose problems: Centralization In a hypothetical scenario, wealthy individuals can buy up governance tokens to accumulate undue influence within the DAO. If the said individual has less-than-benign intentions, the future of the DAO may be threatened. This problem was on full display during the Steem-Tron saga of 2020. has a much more of the dispute, but here's a TL;DR version: Decrypt detailed breakdown Justin Sun, the owner of Tron blockchain, bought social media blockchain Steem. However, community members were unaware of the deal, and understandably, rejected Tron’s takeover of Steem. Now, Steem is a delegated proof-of-stake protocol where token holders can vote on the selection of "Witnesses". Witnesses are "supernodes" responsible for securing the blockchain and validating transactions on the Steem blockchain. [ ] Image source Witnesses have an outsized influence, such that anyone who controls the Witnesses controls the network. Thus, Witness selection became the new battleground for Justin Sun and community members opposing the takeover. In a protracted battle worthy of the Netflix treatment, Sun and hostile Steem holders entered a slugfest to wrest control of the blockchain. Each camp accumulated eye-popping amounts of tokens to secure Witness positions for their preferred representatives. Benefiting from a large war chest, Sun (himself a billionaire) succeeded in electing his Witnesses, effectively taking control of Steem despite overwhelming opposition. Similar problems have surfaced in EOS, another blockchain where 'Delegates' are elected to validate blocks. Certain individuals , and even to use tokens belonging to others, to vote for select representatives. buy tokens collude with exchanges In these cases, token-based governance degenerates into a form of plutocracy where wealthy elites control power. Which goes against everything that on-chain governance and DAOs promise. Rational Indifference If tokens held = voting power, then it follows that DAO members with smaller token holdings may have lower incentives to participate in governance. An individual may see their influence on proposal outcomes as minimal and rationalize refusal to participate in voting. "Rational indifference", as it's known, can discourage DAO members from active participation, leaving token whales at the helm of governance. This re-introduces the principal-agent problem and negates the essential value-proposition for DAOs: putting power in the hands of every stakeholder. Rational indifference may also make collusion easier. With little incentive to vote on proposals, DAO members can sell their governance tokens to the highest bidder for voting purposes. Or, they can "lend" their tokens via flash loans and re-acquire them after the other entity might have used them to . influence voting outcomes Incompetent Leadership In a DAO, policymaking may be crowdsourced and open to anyone holding governance tokens. While this ensures every stakeholder has input in policies, it may give rise to a kakistocracy—leadership by the least competent—especially if the wrong individuals acquire the lion's share of tokens. English author Douglas Adams once quipped: "It is a well-known fact that those people who most want to rule people are, ipso facto, those least suited to do it." With governance tokens circulating on the open market, entities can effectively buy their way into power. Such individuals may not necessarily have the competence or expertise to justify their influence on decision-making. Worse, they may be aligned against the interests of the larger collective in the DAO. For example, a shadowy investor hoping to profit from shorting a DAO's tokens may deliberately destabilize governance mechanisms. The threat of "dark DAOs", malicious groups coordinating to , is extensively covered . exhibit undue influence on DAOs through token accumulation here This problem, known as the Tragedy of Commons, already exists in traditional economies. And it looks like DAOs may be the next victim of the selfish or unwise behavior of a small group. Can DAO Governance Improve? Finding a governance mechanism fit for purpose is a difficult task. The right DAO voting protocol must account for the problems discussed in the preceding section, a task easier said than done. Nonetheless, there are various experiments in the ecosystem to improve the current governance process. This section will discuss some of these alternatives to traditional 1T1V systems and consider their utility: 1. Delegated Voting/Liquid Democracy Sometimes, DAO members can be indifferent to proposals simply because they don't have the mental bandwidth to understand the issues at stake. Especially if the said individuals treat participation in a DAO as a part-time activity, they may be unwilling to commit the time necessary to decide on important matters. To mitigate such problems, delegated voting aka "liquid democracy" is a possible solution. Delegated voting is just that—delegating an individual to vote on your behalf. In this case, the delegate is likely a reputable member of the DAO or an individual with proven expertise on the issues under consideration. Delegated voting using Tally. [ ] Image source Aragon is among the high-profile DAOs to use delegated voting to solve governance-related problems. Liquid democracy encourages competent decision-making, since power is linked to their reputation or expertise, not token holdings. Plus, delegated voting may theoretically produce decisions that reflect the wishes of the community. "But doesn't delegated voting introduce the same issues as coinholder voting?" Delegated voting can indeed increase the possibility of collusion, bribery, and the infamous principal-agent problem. However, well-designed delegated voting systems can be an improvement on existing voting protocols. For starters, DAO members can withdraw their delegated powers anytime they wish. This contrasts with corporate governance mechanisms where removing a rogue Board of Directors can prove difficult. Thus, it becomes near-impossible to coerce individuals or suppress opinions since token holders can choose to exercise their rights. This also keeps delegates accountable to members of the DAO and mitigates the “principal-agent” problem common in traditional governance. 2. Holographic Consensus Holographic consensus is a voting system developed by DAOstack to scale DAO governance. Right now, there are DAOs with thousands of members, and getting members to vote on proposal can be difficult, if not outright impossible. every The thinking behind is that attention spans are limited, so only the most important proposals should come up for voting. Proposals deemed unnecessary can be screened out, ensuring that the attention of DAO members is harnessed properly. holographic consensus Holographic consensus uses a cutting-edge process that combines elements of prediction markets with DAO governance. Here's how that works in practice: 1. Governance is facilitated through DAOstack's collective attention token, GEN. Token holders cannot vote on proposals, but they can 'bet' on which proposals are likely to succeed. If the proposal passes, holders get rewarded with more GEN tokens. But if the proposal falls through, their GEN holdings are slashed. 2. Any DAO member who meets the required specifications can initiate a proposal. Proposals that receive high GEN bets are 'boosted' and pass to the next round. At this stage, proposals with low GEN bets are screened out. 3. Proposals from Round 1 are voted on by members with voting rights. GEN holders can either make profits or losses on their bets depending on whether their choice proposal passes or not. Afterward, approved proposals are executed and implemented in the DAO. [ ] Image source Holographic consensus makes sense for several reasons. It reduces , which occurs when individuals are required to vote too many times. In HC-enabled governance, DAO members can concentrate on high-stake proposals and quickly decide on them. voter fatigue Moreover, it prevents malicious individuals from attacking a DAO by initiating nefarious proposals. A would-be attacker needs to boost a proposal, which means spending their money to buy GEN tokens. That increases the cost of attack and may discourage governance attacks on DAOs. However, there are questions over the perceived usefulness of holographic consensus. For instance, Holographic consensus introduces complexity into DAO governance, and members may find the learning curve too steep. Besides, nothing rules out GEN holders from holding governance tokens. This leaves the door open for bettors to influence the results of the voting. After all, they have everything to lose if their boosted proposal fails to go through. 3. Weighted/Reputation-Based Voting Earlier, we explored the problem of outsiders, such as crypto whales, buying governance tokens to exercise undue influence in a DAO. In this scenario, someone disinterested in the long-term survival of the collective can accrue enough power to destabilize it for selfish gain. To prevent this problem, some DAOs are using weighted and/or reputation-based voting systems. These two concepts are somewhat similar, although differences remain. In weighted voting schemes, the voting power of tokens depends on how long they've been held. Weighted voting schemes incentivize holders to lock up their tokens to improve the utility of their votes. Weighted voting schemes make it harder to acquire voting power in a DAO. You can't just borrow governance tokens from the market and immediately use them to influence voting results. This won't rule out governance attacks entirely, but it significantly increases the cost of executing one. Weighted voting in Kusama parachain. [ ] Image source The follows a similar pattern. Here, DAO members get a single, non-transferable token specifically for voting on proposals. And the voting power attached to each token is linked to the reputation of its holder. reputation-based voting system 'Reputation points' can be earned in different ways. A DAO may reward individuals with reputation points according to their membership history. Or, members can earn more reputation points by contributing positively to the organization's growth (for example, by completing tasks). Reputation voting can ensure individuals with a rich history of participation in the organization acquire great influence. This means an outsider will find it cumbersome to infiltrate the DAO through governance mechanisms. However, reputation voting may be deemed unfair in certain instances. For example, what if a reputable member suddenly goes rogue and uses their influence in the organization for malicious purposes. Wouldn't reputation systems just re-introduce the Tragedy of Commons? Because someone acquired reputation points during the early phase of a project doesn't mean they should hold leadership status forever. Especially if their contribution declines, it'd be unfair for genesis members to wield more power than new but active members. There are two possible solutions here. First, DAOs can make reputation points revocable. If a member is guilty of misconduct, their reputation points should decline to reflect that. Second, reputation points can be programmed to reduce over time, or even expire. This solves the problem of unfairness and keeps power in DAOs fluid. Reputation systems have other flaws, though. Anyone can sell their private keys to transfer reputation (reputation points are linked to wallets). Still, it's difficult to form a thriving market around the sale of reputation points, so this is a viable mechanism for promoting effective DAO governance. 4. Quadratic Voting Quadratic voting is a novel governance mechanism that’s from the likes of Ethereum founder Vitalik Buterin. Quadratic voting is appealing because it offers an alternative to coinholder voting and reduces the influence of whales on voting outcomes. gained support In a quadratic voting scheme, everyone gets one vote once on a proposal. However, they can purchase more “credits” to cast more votes for their preferred option. "Wait, isn't that essentially what 1T1V systems do?" Well, not exactly. In a 1T1V system, the cost of additional votes is linear. If you need five extra votes, you just need five more tokens. In a quadratic voting system, the cost of additional voting is quadratic; one vote costs one token, two votes cost four tokens, three votes cost three tokens, and so on. [ ] Image source Quadratic voting allows individuals to express the extent of their support for a proposal by purchasing vote credits. This makes quadratic voting the DAO version of "putting your money where your mouth is." At the same time, it reduces the influence of crypto whales in decision-making. From our earlier example, it's obvious that the utility of additional voting decreases, while costs increase. It'd take a lot of money for a dark DAO to acquire enough tokens to distort outcomes. Of course, that doesn't rule out vote-buying. For instance, quadratic voting even makes token-selling more lucrative since first votes count for a lot. However, projects like Gitcoin DAO alleviate this problem by using stringent identification systems that make it difficult to transfer voting rights. Users have to register a profile with decentralized identity services like Proof of Humanity, BrightID, Ethereum Naming Service (ENS), and Idena before using their votes. Final Thoughts In its current form, DAO governance is far from perfect. But the sector is investing in novel governance mechanisms to achieve higher member participation, equitable distribution of power, and efficient decision-making. Whether experiments like quadratic voting and holographic consensus will replace traditional coinholder voting is up for debate. Nonetheless, we can expect DAOs to keep reiterating until they find the ideal governance mechanism that promises the most benefits to members and the larger ecosystem. Also Published Here

Stories, News & Mentions about KASISTO

Kasisto’s Stories on HackerNoon