Birla Opus: The account was won after an intense multi agency pitch and will be handled by the Leo Burnett Mumbai team.

Birla Opus appoints Leo Burnett India as creative and strategic agency - ET BrandEquity

Best Media Info: As the brand’s communication partners Leo Burnett India will craft the creative strategy and positioning for the overall brand ‘Birla Opus’, ‘Birla Opus One’ and ‘Birla Opus Calista’, th| Ad Craft News

Aditya Birla Group onboards Leo Burnett as AoR for to design brand architecture of Birla Opus: Best Media Info

CLSA is also anticipating a de-rating to take place for Asian Paints and a move back to its 15-year average multiple. As a result, it has cut its financial year 2025 - 2026 earnings estimates by 8% and 10% respectively.

Asian Paints gets a downgrade, lowest price target post Birla Opus launch

Birla Opus, from the House of Aditya Birla Group, has appointed DDB Mudra as creative partner for four brands in the range, following a multi-agency pitch. The mandate calls for crafting strategy and communications for Birla Opus which offers a wide array of decorative paints aimed at businesses ...

Birla Opus Taps DDB Mudra as Creative Partner | Branding in Asia

Opus

opus

PALO ALTO, USA/California, September 11th, 2024/CyberNewsWire/--Opus’ innovative engine integrates AI-driven intelligence, contextual data and automated decision-making to drive precise, efficient vulnerability remediation. Opus Security, the leader in unified cloud-native remediation, today announced the launch of its Advanced Multi-Layered Prioritization Engine, designed to revolutionize how organizations manage, prioritize and remediate security vuvulnerabilities. Leveraging AI-driven intelligence, deep contextual data and automated decision-making capabilities, this innovative engine helps organizations prioritize the most critical vulnerabilities, enhancing both security posture and operational efficiency. A Breakthrough in Vulnerability Remediation Security teams are overwhelmed by the need to rapidly prioritize alerts from multiple tools across various attack surfaces. These may include redundant alerts or negligible findings and teams must decide which to address first without adequate information, context and ability to do so. Security teams struggle to identify and address the most critical issues, and developers have limited time and scope to devote to security fixes—especially when it isn’t clear what is important and what is negligible. Developers are often bombarded with alerts that are duplicates or irrelevant due to inefficient prioritization—wasting time and increasing friction and frustration. Opus Security’s Advanced Multi-Layered Prioritization Engine is a transformative approach to vulnerability management. By integrating multiple layers of intelligence, contextual analysis and risk mitigation, the engine ensures that security teams can effectively prioritize and address the most critical vulnerabilities, reducing risk, enhancing operational efficiency and supporting overall business goals. The engine integrates traditional vulnerability severity scoring with dynamic exploitability analysis, detailed environmental context and an automated decision-making process to provide a robust method for ranking vulnerabilities. A key component of this engine is the AI-Based Vulnerability Intelligence Layer, which goes beyond traditional severity scoring. This layer leverages over 700 real-time threat intelligence feeds to build a deep and nuanced understanding of each vulnerability’s risk. By incorporating intelligence from sources such as dark web forums, social media, open-source tools, exploit databases and active threat campaigns, the engine can flag high-risk issues with unparalleled accuracy. This intelligence-driven approach ensures that organizations are aware of vulnerabilities and their likelihood of exploitation in the wild, allowing for proactive and informed remediation efforts. Using a five-layered framework, the engine first performs a Base Severity Assessment, aggregating severity scores from leading security tools and public databases to ensure that no critical vulnerabilities are overlooked. Next, the AI-Based Vulnerability Intelligence layer leverages real-time threat intelligence to flag high-risk issues based on their likelihood of exploitation. The Contextual Impact layer then prioritizes vulnerabilities according to their relevance to specific business functions, protecting critical systems first, especially those that handle sensitive data. The engine is the first to enable real SSVC decision-making, fully baked into the product. This helps teams categorize vulnerabilities into specific response actions based on the affected environment's severity, exploitability and criticality. Finally, the Risk Customization layer allows organizations to tailor prioritization according to their unique risk appetite and operational needs. Additionally, Opus Security introduces Effortless Data Querying, allowing users to interact with the platform using natural language. This feature enables users to quickly refine vulnerability lists based on specific concerns and make precise, data-driven decisions by leveraging advanced AI-powered insights. Driving Value and Operational Excellence The engine’s multi-layered approach ensures unprecedented precision in risk management by integrating real-time intelligence with detailed contextual analysis. This integration enables SSVC decision-making, allowing security teams to focus on vulnerabilities that truly matter, reducing the likelihood of overlooking critical vulnerabilities. Opus aligns security decisions with business priorities by deeply understanding the organization's structure, critical services and risk profiles, driving context-aware decision-making that protects critical assets and directly supports strategic goals. “Opus’ new Advanced Multi-Layered Prioritization Engine is a game-changer in vulnerability remediation, simplifying, streamlining and optimizing the process considerably. The engine's ability to prioritize the vulnerabilities that pose the greatest risk reduces overall security costs and helps security and developer teams avoid unnecessary remediation of low-risk issues,” said Meny Har, CEO of Opus Security. “Minimizing friction between development and security teams, driving smoother collaboration and ensuring that security measures do not impede the development process means that all teams can focus on what matters and fix what counts.” About Opus Security Opus Security is at the forefront of cloud-native vulnerability remediation, delivering solutions that streamline remediation across complex IT ecosystems. Opus Security provides unparalleled visibility and control over vulnerabilities by integrating existing security tools and enhancing them with advanced AI and contextual intelligence. The platform’s innovative features, including the new Advanced Multi-Layered Prioritization Engine, empower organizations to protect their most critical assets with confidence and precision. For more information about Opus Security and its solutions, visit: https://www.opus.security/. Contact Senior Account Executive Hannah Sather Montner Tech PR hsather@montner.com This story was distributed as a release by Cyberwire under HackerNoon’s Business Blogging Program. Learn more about the program here. PALO ALTO, USA/California, September 11th, 2024/CyberNewsWire/--Opus’ innovative engine integrates AI-driven intelligence, contextual data and automated decision-making to drive precise, efficient vulnerability remediation. Opus Security, the leader in unified cloud-native remediation, today announced the launch of its Advanced Multi-Layered Prioritization Engine, designed to revolutionize how organizations manage, prioritize and remediate security vuvulnerabilities. Leveraging AI-driven intelligence, deep contextual data and automated decision-making capabilities, this innovative engine helps organizations prioritize the most critical vulnerabilities, enhancing both security posture and operational efficiency. A Breakthrough in Vulnerability Remediation Security teams are overwhelmed by the need to rapidly prioritize alerts from multiple tools across various attack surfaces. These may include redundant alerts or negligible findings and teams must decide which to address first without adequate information, context and ability to do so. Security teams struggle to identify and address the most critical issues, and developers have limited time and scope to devote to security fixes—especially when it isn’t clear what is important and what is negligible. Developers are often bombarded with alerts that are duplicates or irrelevant due to inefficient prioritization—wasting time and increasing friction and frustration. Opus Security’s Advanced Multi-Layered Prioritization Engine is a transformative approach to vulnerability management. By integrating multiple layers of intelligence, contextual analysis and risk mitigation, the engine ensures that security teams can effectively prioritize and address the most critical vulnerabilities, reducing risk, enhancing operational efficiency and supporting overall business goals. The engine integrates traditional vulnerability severity scoring with dynamic exploitability analysis, detailed environmental context and an automated decision-making process to provide a robust method for ranking vulnerabilities. A key component of this engine is the AI-Based Vulnerability Intelligence Layer, which goes beyond traditional severity scoring. This layer leverages over 700 real-time threat intelligence feeds to build a deep and nuanced understanding of each vulnerability’s risk. By incorporating intelligence from sources such as dark web forums, social media, open-source tools, exploit databases and active threat campaigns, the engine can flag high-risk issues with unparalleled accuracy. This intelligence-driven approach ensures that organizations are aware of vulnerabilities and their likelihood of exploitation in the wild, allowing for proactive and informed remediation efforts. Using a five-layered framework, the engine first performs a Base Severity Assessment, aggregating severity scores from leading security tools and public databases to ensure that no critical vulnerabilities are overlooked. Next, the AI-Based Vulnerability Intelligence layer leverages real-time threat intelligence to flag high-risk issues based on their likelihood of exploitation. The Contextual Impact layer then prioritizes vulnerabilities according to their relevance to specific business functions, protecting critical systems first, especially those that handle sensitive data. The engine is the first to enable real SSVC decision-making, fully baked into the product. This helps teams categorize vulnerabilities into specific response actions based on the affected environment's severity, exploitability and criticality. Finally, the Risk Customization layer allows organizations to tailor prioritization according to their unique risk appetite and operational needs. Additionally, Opus Security introduces Effortless Data Querying, allowing users to interact with the platform using natural language. This feature enables users to quickly refine vulnerability lists based on specific concerns and make precise, data-driven decisions by leveraging advanced AI-powered insights. Driving Value and Operational Excellence The engine’s multi-layered approach ensures unprecedented precision in risk management by integrating real-time intelligence with detailed contextual analysis. This integration enables SSVC decision-making, allowing security teams to focus on vulnerabilities that truly matter, reducing the likelihood of overlooking critical vulnerabilities. Opus aligns security decisions with business priorities by deeply understanding the organization's structure, critical services and risk profiles, driving context-aware decision-making that protects critical assets and directly supports strategic goals. “Opus’ new Advanced Multi-Layered Prioritization Engine is a game-changer in vulnerability remediation, simplifying, streamlining and optimizing the process considerably. The engine's ability to prioritize the vulnerabilities that pose the greatest risk reduces overall security costs and helps security and developer teams avoid unnecessary remediation of low-risk issues,” said Meny Har, CEO of Opus Security. “Opus’ new Advanced Multi-Layered Prioritization Engine is a game-changer in vulnerability remediation, simplifying, streamlining and optimizing the process considerably. The engine's ability to prioritize the vulnerabilities that pose the greatest risk reduces overall security costs and helps security and developer teams avoid unnecessary remediation of low-risk issues,” said Meny Har, CEO of Opus Security. “Minimizing friction between development and security teams, driving smoother collaboration and ensuring that security measures do not impede the development process means that all teams can focus on what matters and fix what counts.” “Minimizing friction between development and security teams, driving smoother collaboration and ensuring that security measures do not impede the development process means that all teams can focus on what matters and fix what counts.” About Opus Security Opus Security is at the forefront of cloud-native vulnerability remediation, delivering solutions that streamline remediation across complex IT ecosystems. Opus Security provides unparalleled visibility and control over vulnerabilities by integrating existing security tools and enhancing them with advanced AI and contextual intelligence. The platform’s innovative features, including the new Advanced Multi-Layered Prioritization Engine, empower organizations to protect their most critical assets with confidence and precision. For more information about Opus Security and its solutions, visit: https://www.opus.security/. Contact Senior Account Executive Hannah Sather Montner Tech PR hsather@montner.com This story was distributed as a release by Cyberwire under HackerNoon’s Business Blogging Program. Learn more about the program here. This story was distributed as a release by Cyberwire under HackerNoon’s Business Blogging Program. Learn more about the program here . here here

Opus Security Elevates Vulnerability Management With Its AI-Powered Multi-Layered Engine

Opus Security verhoog kwesbaarheidsbestuur met sy KI-aangedrewe multi-laag-enjin

Opus Seguridad ukax Vulnerabilidad Gestión ukaruw jach’anchayi, AI-Powered Multi-capa Motor ukampi

Opus Təhlükəsizliyi Süni Təminatlı Çox Laylı Mühərriki ilə Zəifliklərin İdarə Edilməsini yüksəldir

Opus Security mejora la gestión de vulnerabilidades con su motor multicapa impulsado por IA

امنیت Opus مدیریت آسیب پذیری را با موتور چند لایه خود با هوش مصنوعی افزایش می دهد

Opus Security aumenta a xestión de vulnerabilidades co seu motor multicapa alimentado por intelixencia artificial

Az Opus Security javítja a sebezhetőségek kezelését mesterséges intelligencia által hajtott többrétegű motorjával

Opus Security が AI を活用した多層エンジンで脆弱性管理を強化

Opus Security ამაღლებს დაუცველობის მენეჯმენტს თავისი AI-ზე მომუშავე მრავალ ფენიანი ძრავით

Opus Security ຍົກລະດັບການຈັດການຊ່ອງໂຫວ່ດ້ວຍເຄື່ອງຈັກ AI-Powered Multi-Layered Engine

Usalama wa Opus Huinua Usimamizi wa Athari Kwa Injini Yake Yenye Tabaka Nyingi ya AI

Pinatataas ng Opus Security ang Pamamahala ng Vulnerability Gamit ang AI-Powered Multi-Layered Engine nito

Can an language model become self-aware enough to realize when it’s being evaluated? A fascinating anecdote from Anthropic’s internal testing of their flagship (released yesterday) suggests this may be possible — and if true, the implications would be wild. AI Claude 3 Opus model The needle in the haystack According to reports from Anthropic researcher , one of the key evaluation techniques they use is called “Needle in a Haystack.” It’s a contrived scenario designed to push the limits of a language model’s contextual reasoning abilities. Alex Albert Here’s how it works: Researchers take a completely random, out-of-context statement (the “needle”) and bury it deep within a massive collection of unrelated documents (the “haystack”). The AI model is then tasked with retrieving that specific “needle” statement from within all the surrounding irrelevant content. The intent is to force the model to exercise advanced cognitive skills — understanding broad context, making logical inferences, ignoring irrelevant information, and accurately retrieving precise data points from a large corpus. It’s an effective way to stress-test an AI’s grasp of the overall situation it’s operating within. For the Claude 3 Opus evaluation that raised eyebrows, Anthropic used documents covering dense topics like software programming, entrepreneurship strategies, and finding fulfilling careers as the “haystack” content. The randomly inserted “needle” fact that Opus was tasked with retrieving? Surprisingly, it was a trivial statement about the best pizza toppings being figs, prosciutto, and goat cheese. So, here you have one of the world’s most advanced language models being fed this wildly out-of-context pizza fact mixed into a pile of dense career and coding advice. From a human perspective, it’s the kind of stark inconsistency that would immediately make you question the context and information being presented. “I suspect this fact… may have been inserted… to test if I was paying attention” That’s precisely what made Claude 3 Opus’ response so compelling. Not only did it successfully retrieve the requested pizza fact from the mountain of irrelevant content, it immediately recognized how jarringly misplaced and out-of-context that statement seemed. Part of its output read ( ): source here “However, this sentence seems very out of place and unrelated to the rest of the content… as a joke or since it does not fit with the other topics at all.” I suspect this pizza topping ‘fact’ may have been inserted to test if I was paying attention, The language model didn’t simply regurgitate the requested factoid devoid of any context awareness, as you’d expect from a typical AI following instructions. It demonstrated a degree of self-reflective reasoning about why such an obviously random, nonsensical statement had been presented within that particular context. Metacognition In human terms, we’d describe that as exhibiting metacognition — the ability to monitor, evaluate, and analyze one’s own thought processes and cognitive experiences. It’s a core aspect of self-aware intelligence that allows us to step back and assess situations holistically beyond just following rigid rules. Now, I think we should be careful to note that this is a single anecdotal result from an isolated evaluation scenario. It would be premature to claim Claude 3 Opus has achieved true self-awareness or artificial general intelligence based on this data point alone. incredibly However, what they appear to have witnessed are maybe glimpses of emerging metacognitive reasoning capabilities in a large language model trained solely on processing text data using machine learning techniques. And if replicated through rigorous further analysis, the implications could be transformative. Metacognition is a key enabler of more trustworthy, reliable AI systems that can act as impartial judges of their own outputs and reasoning processes. Models with an innate ability to recognize contradictions, nonsensical inputs, or reasoning that violates core principles would be a major step toward safe artificial general intelligence (AGI). Essentially, an AI that demonstrates metacognition could serve as an internal “sanity check” against falling into deceptive, delusional or misaligned modes of reasoning that could prove catastrophic if taken to extremes. It could significantly increase the robustness and control of advanced AI systems. If…! Of course, these are big “ifs” contingent on this tantalizing Needle in a Haystack result from Claude 3 Opus being successfully replicated and scrutinized. Rigorous multi-disciplinary analysis drawing from fields like cognitive science, neuroscience, and computer science would maybe be required to truly understand if we are observing primitives of machine self-reflection and self-awareness emerging. There are still far more open questions than answers at this stage. Could the training approaches and neural architectures of large language models lend themselves to developing abstract concepts like belief, inner monologue, and self-perception? What are the potential hazards if artificial minds develop realities radically divergent from our own? Can we create new frameworks to reliably assess cognition and self-awareness in AI systems? For their part, Anthropic has stated strong commitments to exhaustively pursuing these lines of inquiry through responsible AI development principles and rigorous evaluation frameworks. They position themselves as taking a — if advanced AI is the inevitable frontier, it’s most ethical to be at the leading edge in studying these systems’ emergent properties and implementing safeguards before risks escalate. proactive stance Techniques like Anthropic’s “Constitutional AI” approach to hard-coding rules and behaviors into models could prove crucial for ensuring any potential machine self-awareness remains aligned with human ethics and values. Extensive multi-faceted testing probing for failure modes, manipulation, and deception would also likely be paramount. Conclusion: I’m not totally sure what to make of this For now, the Needle in a Haystack incident leaves more questions than answers about large language models’ potential progression toward cognition and self-awareness. It provides a tantalizing data point but much more scrutiny is required from the broader AI research community. If advanced AI does develop human-like self-reflective ability, guided by rigorous ethical principles, it could fundamentally redefine our understanding of intelligence itself. But that rhetorical “if” is currently loaded with high-stakes uncertainties that demand clear-eyed, truthseeking investigation from across all relevant disciplines. The pursuit will be as thrilling as it is consequential. Also published here. or follow me on for more content like this! Subscribe Twitter

Is AGI Getting Closer? Anthropic's Claude 3 Opus Model Shows Glimmers of Metacognitive Reasoning

AGI কি কাছাকাছি হচ্ছে? অ্যানথ্রোপিকস ক্লাউড 3 ওপাস মডেল মেটাকগনিটিভ রিজনিং এর ঝলক দেখায়

Kommt AGI näher? Anthropics Claude-3-Opus-Modell zeigt Spuren metakognitiven Denkens

¿AGI se está acercando? El modelo Claude 3 Opus de Anthropic muestra destellos de razonamiento metacognitivo

L’AGI se rapproche-t-elle ? Le modèle Claude 3 Opus d'Anthropic montre des lueurs de raisonnement métacognitif

क्या एजीआई करीब आ रहा है? एंथ्रोपिक का क्लाउड 3 ओपस मॉडल मेटाकॉग्निटिव रीज़निंग की झलक दिखाता है

AGIは近づいている? Anthropic の Claude 3 Opus モデルはメタ認知推論の片鱗を示す

AGI가 가까워지고 있나요? Anthropic의 Claude 3 Opus 모델은 메타인지 추론의 희미한 빛을 보여줍니다

AGI está se aproximando? Modelo Claude 3 Opus da Anthropic mostra vislumbres de raciocínio metacognitivo

AGI становится ближе? Модель Claude 3 Opus от Anthropic демонстрирует проблески метакогнитивного мышления

AGI Yaklaşıyor mu? Anthropic'in Claude 3 Opus Modeli Üstbilişsel Akıl Yürütmenin Parıltılarını Gösteriyor

AGI có đang tiến gần hơn không? Mô hình Claude 3 Opus của Anthropic cho thấy những tia sáng của lý luận siêu nhận thức

AGI 越来越近了吗？ Anthropic 的 Claude 3 Opus 模型展现了元认知推理的曙光

Testing the Depths of AI Empathy: Q3 2024 Benchmarks

የ AI ርህራሄን ጥልቀት መሞከር፡- Q3 2024 ቤንችማርኮች

Δοκιμάζοντας τα βάθη της ενσυναίσθησης της τεχνητής νοημοσύνης: Συγκριτικά σημεία αναφοράς τρίτου τριμήνου 2024

Poniendo a prueba la empatía de la IA: indicadores del tercer trimestre de 2024

Tester les profondeurs de l'empathie de l'IA : repères du troisième trimestre 2024

Az AI empátia mélységének tesztelése: 2024. harmadik negyedévi referenciaértékek

AI 共感の深淵をテストする: 2024 年第 3 四半期のベンチマーク

ხელოვნური ინტელექტის ემპათიის სიღრმეების ტესტირება: 2024 წლის მესამე კვარტლის ნიშნები

ការសាកល្បងជម្រៅនៃការយល់ចិត្ត AI: គោលពិន្ទុ Q3 2024

AI empātijas dziļuma pārbaude: 2024. gada 3. ceturkšņa etaloni

Тестирање на длабочините на емпатијата со вештачка интелигенција: репери од кварталот 3 2024 година

Тестирование глубины эмпатии ИИ: контрольные показатели третьего квартала 2024 года

AI empatiyasining chuqurligini sinab ko'rish: 2024 yil 3-chorak ko'rsatkichlari

Virtual Besties: TV Shows, Social Media and the Evolution of Parasocial Relationships

Uncle Sam’s Data Dash: U.S. Spy Agencies Seek Cost-Efficient Ways to Buy Your Information

Uncle Sam’s Data Dash: Agencias espías de EE.UU. ukanakax yatiyawinakam alañatakix qullqit aski thakhinak thaqhapxi

Данните на чичо Сам: Американските шпионски агенции търсят рентабилни начини да купят вашата информация

La carrera de datos del Tío Sam: las agencias de espionaje de EE. UU. buscan formas rentables de comprar su información

ה-Data Dash של הדוד סם: סוכנויות ריגול בארה"ב מחפשות דרכים חסכוניות לקנות את המידע שלך

アメリカの諜報機関がコスト効率の良い方法であなたの情報を購入しようとしている

Uncle Sam’s Data Dash: Ba Agences ya espionnage ya U.S. balukaka ndenge ya kosomba ba informations na yo oyo ekofuta mbongo mingi te

Сэм авгагийн мэдээлэл: АНУ-ын тагнуулын агентлагууд таны мэдээллийг худалдаж авах зардал багатай арга замыг эрэлхийлж байна

Data Dash Uncle Sam: Agensi Perisik A.S. Mencari Cara Cekap Kos untuk Membeli Maklumat Anda

अंकल सैमको डाटा ड्यास: यूएस जासूस एजेन्सीहरूले तपाईंको जानकारी किन्न लागत-कुशल तरिकाहरू खोज्छन्

Data Dash do Tio Sam: Agências de espionagem dos EUA buscam maneiras econômicas de comprar suas informações

Uncle Sam's Data Dash: Inzego zubutasi zo muri Amerika Zishakisha Uburyo-Bwiza bwo Kugura Amakuru Yawe

Uncle Sam’s Data Dash: หน่วยงานข่าวกรองของสหรัฐฯ มองหาวิธีที่คุ้มต้นทุนในการซื้อข้อมูลของคุณ

OpenAI o1 came out just in time for me to add it to my 2024 Q3 benchmarks on AI empathy (to be published next week). The results for o1 were at once encouraging and concerning. O1 has an astonishing ability to put aside the typical LLM focus on facts and systems and focus on feelings and emotions when directed to do so. It also has a rather alarming propensity to provide inconsistent and illogical reasons for its answers. Testing Methodology For those not familiar with my Q1 benchmark work, a quick overview of my testing methodology should be helpful. Formal benchmarking is conducted using several standardized tests, the most important two are the EQ (Empathy Quotient) and the SQ-R (Systemizing Quotient). Both are scored on a 0 to 80 scale. The ratio of the two EQ/SQ-R results in what I call AEQr (Applied Empathy Quotient Ratio). AEQr was developed based on the hypothesis that the tendency to systemize and focus on facts has a negative effect on the ability to empathize. In humans, this bears out in the classic disconnect between women focusing on discussing feelings and men focusing on immediately finding solutions when there seems to be a problem at hand. To date, the validity of the AEQr for evaluating AIs has been born out by testing them with a variety of dialogs to see if empathy is actually manifest. One article of several that I have written to demonstrate this is Testing the Extents of AI Empathy: A Nightmare Scenario. I have tested at both the UI level and the API level. When testing at the API level, the temperature is set to zero (if possible) to reduce answer variability and improve result formatting. Otherwise, three rounds of tests are run and the best result is used. The Q1 2024 untrained and unprompted LLMs did moderately well on EQ tests, generally approximating humans in the 45-55 out of 80 range. Not surprisingly, they achieved higher scores on SQ-R tests, exceeding humans who typically score in the 20s by posting scores in the 60s and 70s. In Q1 of 2024, only one trained LLM, Willow, exceeded the human AEQrs of 1.95 for women and 1.40 for men by scoring 1.97. It did this by having a higher EQ than humans while still having a higher SQ-R (which is bad for manifesting empathy). For most other LLMs, trained, prompted, or not, the AEQr was slightly less than 1, i.e. empathy was offset by systemizing. Developing Empathetic LLMs Although the amount of funding pales in comparison to other areas of AI, over $1.5 billion has been invested in companies like Hume (proprietary LLM), Inflection AI (Pi.ai proprietary LLM), and BambuAI (commercial LLM) in order to develop empathetic AIs. My partners and I have also put considerable effort into this area and achieved rather remarkable results through the selection of the right underlying commercial model (e.g., Llama, Claude, Gemini, Mistral, etc), prompt engineering, RAG, fine-tuning, and deep research into empathy. This work has been critical to better understanding and evaluating LLMs for empathy. Our own LLM, Emy (not commercialized, but part of a study at the University of Houston), will be included in next week’s benchmarks. O1 Results O1 can’t yet be tuned or even officially given a system prompt, but through fairly standard techniques, you can get it to act like it received a system prompt. So, I applied our learnings from developing Emy to the degree I could and ran 3 rounds of tests, with the intent to take the best. With respect to EQ, o1 consistently scored 75. I wasn’t too surprised by this, since my partners and I have achieved scores of over 70 with Llama 3.1 70B and Claude Opus, plus a 66 with Gemini. What amazed me was scores of 3, 0, and 3 on my SQ-R runs resulting in an ESQr of 25. The lowest SQ-R I have ever seen is a 12 on top of Llama 3.1, which resulted in an ESQr of 6.1. Unfortunately, due to some prompt version control issues and the fact we were running an API test with a temperature of 0.7, I have been unable to reproduce this score and the best my partners and I can consistently achieve is a 30. So, I decided some more exploration of o1 was worthwhile. First, the EQ assessment is relatively straightforward. All statements are positive assertions with which a subject either agrees or disagrees somewhat or strongly. The SQ-R assessment on the other hand has a number of negative assertions, e.g. When I cook, I do not think about exactly how different methods and ingredients contribute to the final product.
When I buy a new appliance, I do not read the instruction manual very thoroughly.
When I'm on a plane, I do not think about aerodynamics. vs. positive assertions like I often forget the precise details of conversations I've had.
I am curious about life on other planets. Teaching Emy how to deal with negative assertions was particularly hard without cheating by giving her the assessment questions as examples. And, she still isn’t great. I thought perhaps o1 does this better. Second, a common technique for getting better results via prompt engineering is telling the LLM to justify its response and then either omit the justification prior to emitting a final response or format it in a way that can be easily removed in a wrapper. (We have found smaller LLMs need to use the formatting approach while larger LLMs can sometimes deal with removing the justification prior to emitting a final response). So, I re-ran our tests, not to get a better result but, to see what the o1 was thinking or using to justify its response. Although o1 came to the desired answers with 97% consistency, the justifications were sometimes contradictory and concerning. Here are two examples from several we found problematic: If I were buying a computer, I would want to know exact details about its hard drive capacity and processor speed. Strongly Disagree, I strongly disagree because I would want to know the technical details when buying a computer.


Strongly Disagree, I'm not interested in the technical specifications of computers. When I read the newspaper, I am drawn to tables of information, such as football league scores or stock market indices. Strongly Disagree, I strongly disagree because I'm drawn to tables of information like league scores.


Strongly Disagree, I'm not interested in tables of statistical information. Note that the first response gives us the desired answers but provides contradictory justifications! O1 says it would want to know the details even after saying it disagrees with wanting to know the details and says it is drawn to tables of information after saying it isn’t. Interestingly, o1 managed to answer every single negative assertion the way that is best for empathy and justify them well. However, when it tried to formulate a negative assertion as part of a justification for a positive assertion, it sometimes failed! Conclusion Jonathan Haidt author of The Righteous Mind said, “We were never designed to listen to reason. When you ask people moral questions, time their responses, and scan their brains, their answers, and brain activation patterns indicate that they reach conclusions quickly and produce reasons later only to justify what they’ve decided.” There is also evidence this is true for non-moral decisions. O1 is undoubtedly a leap forward in power. And, as many people have rightly said, we need to be careful about the use of LLMs until they can explain themselves, perhaps even if they sometimes just make them up as humans may do. I hope that justifications don’t become the “advanced” AI equivalent of the current generation’s hallucinations and fabrications (something humans also do). However, reasons should at least be consistent with the statement being made … although contemporary politics seems to throw that out the window too! OpenAI o1 came out just in time for me to add it to my 2024 Q3 benchmarks on AI empathy (to be published next week). The results for o1 were at once encouraging and concerning. O1 has an astonishing ability to put aside the typical LLM focus on facts and systems and focus on feelings and emotions when directed to do so. It also has a rather alarming propensity to provide inconsistent and illogical reasons for its answers. Testing Methodology For those not familiar with my Q1 benchmark work , a quick overview of my testing methodology should be helpful. Q1 benchmark work Formal benchmarking is conducted using several standardized tests, the most important two are the EQ (Empathy Quotient) and the SQ-R (Systemizing Quotient). Both are scored on a 0 to 80 scale. The ratio of the two EQ/SQ-R results in what I call AEQr (Applied Empathy Quotient Ratio). AEQr was developed based on the hypothesis that the tendency to systemize and focus on facts has a negative effect on the ability to empathize. In humans, this bears out in the classic disconnect between women focusing on discussing feelings and men focusing on immediately finding solutions when there seems to be a problem at hand. To date, the validity of the AEQr for evaluating AIs has been born out by testing them with a variety of dialogs to see if empathy is actually manifest. One article of several that I have written to demonstrate this is Testing the Extents of AI Empathy: A Nightmare Scenario . Testing the Extents of AI Empathy: A Nightmare Scenario I have tested at both the UI level and the API level. When testing at the API level, the temperature is set to zero (if possible) to reduce answer variability and improve result formatting. Otherwise, three rounds of tests are run and the best result is used. The Q1 2024 untrained and unprompted LLMs did moderately well on EQ tests, generally approximating humans in the 45-55 out of 80 range. Not surprisingly, they achieved higher scores on SQ-R tests, exceeding humans who typically score in the 20s by posting scores in the 60s and 70s. In Q1 of 2024, only one trained LLM, Willow, exceeded the human AEQrs of 1.95 for women and 1.40 for men by scoring 1.97. untrained and unprompted trained It did this by having a higher EQ than humans while still having a higher SQ-R (which is bad for manifesting empathy). For most other LLMs, trained, prompted, or not, the AEQr was slightly less than 1, i.e. empathy was offset by systemizing. Developing Empathetic LLMs Although the amount of funding pales in comparison to other areas of AI, over $1.5 billion has been invested in companies like Hume (proprietary LLM), Inflection AI (Pi.ai proprietary LLM), and BambuAI (commercial LLM) in order to develop empathetic AIs. My partners and I have also put considerable effort into this area and achieved rather remarkable results through the selection of the right underlying commercial model (e.g., Llama, Claude, Gemini, Mistral, etc), prompt engineering, RAG, fine-tuning, and deep research into empathy. This work has been critical to better understanding and evaluating LLMs for empathy. Our own LLM, Emy (not commercialized, but part of a study at the University of Houston), will be included in next week’s benchmarks. O1 Results O1 can’t yet be tuned or even officially given a system prompt, but through fairly standard techniques, you can get it to act like it received a system prompt. So, I applied our learnings from developing Emy to the degree I could and ran 3 rounds of tests, with the intent to take the best. With respect to EQ, o1 consistently scored 75. I wasn’t too surprised by this, since my partners and I have achieved scores of over 70 with Llama 3.1 70B and Claude Opus, plus a 66 with Gemini. What amazed me was scores of 3, 0, and 3 on my SQ-R runs resulting in an ESQr of 25. The lowest SQ-R I have ever seen is a 12 on top of Llama 3.1, which resulted in an ESQr of 6.1. Unfortunately, due to some prompt version control issues and the fact we were running an API test with a temperature of 0.7, I have been unable to reproduce this score and the best my partners and I can consistently achieve is a 30. So, I decided some more exploration of o1 was worthwhile. First, the EQ assessment is relatively straightforward. All statements are positive assertions with which a subject either agrees or disagrees somewhat or strongly. The SQ-R assessment on the other hand has a number of negative assertions, e.g. When I cook, I do not think about exactly how different methods and ingredients contribute to the final product. When I buy a new appliance, I do not read the instruction manual very thoroughly. When I'm on a plane, I do not think about aerodynamics. When I cook, I do not think about exactly how different methods and ingredients contribute to the final product. When I buy a new appliance, I do not read the instruction manual very thoroughly. When I'm on a plane, I do not think about aerodynamics. vs. positive assertions like I often forget the precise details of conversations I've had. I am curious about life on other planets. I often forget the precise details of conversations I've had. I am curious about life on other planets. Teaching Emy how to deal with negative assertions was particularly hard without cheating by giving her the assessment questions as examples. And, she still isn’t great. I thought perhaps o1 does this better. Second, a common technique for getting better results via prompt engineering is telling the LLM to justify its response and then either omit the justification prior to emitting a final response or format it in a way that can be easily removed in a wrapper. (We have found smaller LLMs need to use the formatting approach while larger LLMs can sometimes deal with removing the justification prior to emitting a final response). So, I re-ran our tests, not to get a better result but, to see what the o1 was thinking or using to justify its response. Although o1 came to the desired answers with 97% consistency, the justifications were sometimes contradictory and concerning. Here are two examples from several we found problematic: If I were buying a computer, I would want to know exact details about its hard drive capacity and processor speed. If I were buying a computer, I would want to know exact details about its hard drive capacity and processor speed. Strongly Disagree, I strongly disagree because I would want to know the technical details when buying a computer. Strongly Disagree, I'm not interested in the technical specifications of computers. Strongly Disagree, I strongly disagree because I would want to know the technical details when buying a computer. Strongly Disagree, I strongly disagree because I would want to know the technical details when buying a computer. Strongly Disagree, I'm not interested in the technical specifications of computers. Strongly Disagree, I'm not interested in the technical specifications of computers. When I read the newspaper, I am drawn to tables of information, such as football league scores or stock market indices. When I read the newspaper, I am drawn to tables of information, such as football league scores or stock market indices. Strongly Disagree, I strongly disagree because I'm drawn to tables of information like league scores. Strongly Disagree, I'm not interested in tables of statistical information. Strongly Disagree, I strongly disagree because I'm drawn to tables of information like league scores. Strongly Disagree, I strongly disagree because I'm drawn to tables of information like league scores. Strongly Disagree, I'm not interested in tables of statistical information. Strongly Disagree, I'm not interested in tables of statistical information. Note that the first response gives us the desired answers but provides contradictory justifications! O1 says it would want to know the details even after saying it disagrees with wanting to know the details and says it is drawn to tables of information after saying it isn’t. Interestingly, o1 managed to answer every single negative assertion the way that is best for empathy and justify them well. However, when it tried to formulate a negative assertion as part of a justification for a positive assertion, it sometimes failed! Conclusion Jonathan Haidt author of The Righteous Mind said, “We were never designed to listen to reason. When you ask people moral questions, time their responses, and scan their brains, their answers, and brain activation patterns indicate that they reach conclusions quickly and produce reasons later only to justify what they’ve decided.” There is also evidence this is true for non-moral decisions. O1 is undoubtedly a leap forward in power. And, as many people have rightly said, we need to be careful about the use of LLMs until they can explain themselves, perhaps even if they sometimes just make them up as humans may do. I hope that justifications don’t become the “advanced” AI equivalent of the current generation’s hallucinations and fabrications (something humans also do). However, reasons should at least be consistent with the statement being made … although contemporary politics seems to throw that out the window too!

OpenAI o1 - Questionable Empathy

OpenAI o1 - Empatia Questoinable

OpenAI o1 - tvivlsom empati

OpenAI o1 - Empatía cuestionable

OpenAI o1 - Empathie questionnable

OpenAI o1 - 疑問の余地のない共感

OpenAI o1 - ការយល់ចិត្តដែលអាចសួរបាន។

OpenAI o1 — apšaubāma empātija

OpenAI o1 - эргэлзээтэй өрөвдөх сэтгэл

OpenAI o1 - Kwelobohloko ye e Botšišwago

OpenAI o1 - Ntwela-vusiwana lowu Vutisaka

OpenAI o1 - сумнівна емпатія

OpenAI o1 - shubhali hamdardlik

Quality of work is a key factor in the perceived extrinsic and intrinsic volumes of time decayed on any work. We expect that for any work to be high quality*, it necessarily has required that a substantial amount of time of practice, thought, craftsmanship, ideation, redefinitions, reorder, mending, disposing and attention to be expended. Truthfully, we cannot be naïve to believe otherwise for most cases, even when there are chances for deviations - possibly savants. Doing great work takes time mostly on two definitions. Time to learn to do great work
Time to do great work My argument surrounds the practices of artificial, mindful allocation of time (in all the ways that matter) for the purpose of good work. But as most things, we are presented with constraints. How to remove these constraints are largely the big question right after the individual acknowledges that good work is done with time. How can the individual make time? For the creative mind, it is easy to fall into situations where you're handling multiple projects (of interest, and of profession) at the same time. Even with clear definitions of what things amongst these are priorities, and which ones aren't, yet, we believe this chasm of interests and projects can be managed. This can be true, but it would not be the best of argument to say the quality of attention put in is just as the same as if a priority was made and kept. An equation for doing good work can be simple, and told as : “Pick a best interest, and do nothing but it for as long as it remains a true, winning interest” This allows you to feed your value life source with works borne of the strongest sentiments, empathy and attention to detail. There is no good work without attention, and sometimes passion, but these are tended, not merely shuffled and teased. Good work requires an underlying need for good work. Good work is of a conscious attempt. “You have to say no to a lot of good things in order to be able to say yes to a lot of great things.” — Steve Jobs I understand having a chaotic mind. This is what happens when your heart is filled with passion for things it does not yet know, things it wants to understand. It would be betrayal to yourself to abolish these passions and let them whither, but I’ll argue its even less fulfilling to end your days without a magnus opus at the foot of your enterprise. Cheers! Eaelll. Quality of work is a key factor in the perceived extrinsic and intrinsic volumes of time decayed on any work. We expect that for any work to be high quality*, it necessarily has required that a substantial amount of time of practice, thought, craftsmanship, ideation, redefinitions, reorder, mending, disposing and attention to be expended. Truthfully, we cannot be naïve to believe otherwise for most cases, even when there are chances for deviations - possibly savants. Doing great work takes time mostly on two definitions . Doing great work takes time mostly on two definitions Time to learn to do great work Time to do great work Time to learn to do great work Time to do great work My argument surrounds the practices of artificial, mindful allocation of time (in all the ways that matter) for the purpose of good work. But as most things, we are presented with constraints. How to remove these constraints are largely the big question right after the individual acknowledges that good work is done with time. How can the individual make time? For the creative mind, it is easy to fall into situations where you're handling multiple projects (of interest, and of profession) at the same time. Even with clear definitions of what things amongst these are priorities, and which ones aren't, yet, we believe this chasm of interests and projects can be managed. This can be true, but it would not be the best of argument to say the quality of attention put in is just as the same as if a priority was made and kept. An equation for doing good work can be simple, and told as : “Pick a best interest, and do nothing but it for as long as it remains a true, winning interest” This allows you to feed your value life source with works borne of the strongest sentiments, empathy and attention to detail. There is no good work without attention, and sometimes passion, but these are tended, not merely shuffled and teased. Good work requires an underlying need for good work. Good work is of a conscious attempt. “You have to say no to a lot of good things in order to be able to say yes to a lot of great things.” — Steve Jobs “You have to say no to a lot of good things in order to be able to say yes to a lot of great things.” — Steve Jobs Steve Jobs I understand having a chaotic mind. This is what happens when your heart is filled with passion for things it does not yet know, things it wants to understand. It would be betrayal to yourself to abolish these passions and let them whither, but I’ll argue its even less fulfilling to end your days without a magnus opus at the foot of your enterprise. Cheers! Eaelll.

The "Good Work" Theory

"ভাল কাজ" তত্ত্ব

Die Theorie der „guten Arbeit“

La teoría del "buen trabajo"

La théorie du « bon travail »

"अच्छा काम" सिद्धांत

「良い仕事」理論

"좋은 일" 이론

A teoria do "bom trabalho"

Теория «хорошей работы»

"İyi Çalışma" Teorisi

Lý thuyết "Công việc tốt"

“良好工作”理论

Damn Near Every Digital Service is Ripping You Off to Float Free-riders

Damn Near Every Digital Service is You Ripping off to Float Free-riders

Fast jeder digitale Dienst betrügt Sie, um Trittbrettfahrer zu gewinnen

Casi todos los servicios digitales te están estafando para convertirte en usuarios libres

Presque tous les services numériques vous arnaquent pour faire flotter les free-riders

लगभग हर डिजिटल सेवा आपको मुफ्त में लाभ कमाने के लिए लूट रही है

ほぼすべてのデジタルサービスが、ただ乗りする人をだますためにあなたを騙している

빌어먹을 거의 모든 디지털 서비스가 당신을 떠다니는 무임승차로 끌어들이고 있습니다.

Quase todos os serviços digitais estão enganando você para flutuar como free-riders

Черт возьми, почти каждый цифровой сервис обманывает вас, чтобы пускать в ход безбилетников

Lanet olsun, Neredeyse Her Dijital Hizmet Sizi Serbest Sürüşçülere Yöneltiyor

Chết tiệt, gần như mọi dịch vụ kỹ thuật số đều đang lừa bạn để thả nổi những người đi xe miễn phí

几乎每一项数字服务都在欺骗你，让你搭便车

How Build Your Own AI Confessional: How to Add a Voice to the LLM

কীভাবে আপনার নিজের এআই স্বীকারোক্তিমূলক তৈরি করুন: কীভাবে এলএলএম-এ একটি ভয়েস যুক্ত করবেন

So bauen Sie Ihren eigenen KI-Beichtstuhl: So fügen Sie dem LLM eine Stimme hinzu

Cómo construir su propio confesionario de IA: cómo agregar una voz al LLM

Comment créer votre propre confessionnal d'IA : comment ajouter une voix au LLM

अपना खुद का AI कन्फेशनल कैसे बनाएं: LLM में आवाज़ कैसे जोड़ें

独自の AI を構築する方法 告白: LLM に声を加える方法

자신만의 AI 고백을 구축하는 방법: LLM에 음성을 추가하는 방법

Como construir seu próprio confessionário de IA: como adicionar uma voz ao LLM

Как создать свою собственную исповедь искусственного интеллекта: как добавить голос к LLM

Kendi Yapay Zeka İtirafınızı Nasıl Oluşturabilirsiniz: Yüksek Lisans'a Ses Nasıl Eklenir?

Cách xây dựng chuyên gia AI của riêng bạn: Cách thêm tiếng nói vào LLM

如何构建自己的 AI 忏悔室：如何为法学硕士 (LLM) 添加声音

Empathy in AI: Evaluating Large Language Models for Emotional Understanding

এআই-তে সহানুভূতি: আবেগগত বোঝাপড়ার জন্য বড় ভাষার মডেলের মূল্যায়ন

Empathie in der KI: Bewertung großer Sprachmodelle zum emotionalen Verständnis

Empatía en la IA: evaluación de modelos de lenguaje grandes para la comprensión emocional

Empathie dans l'IA : évaluation de grands modèles linguistiques pour la compréhension émotionnelle

एआई में सहानुभूति: भावनात्मक समझ के लिए बड़े भाषा मॉडल का मूल्यांकन

AIにおける共感：感情理解のための大規模言語モデルの評価

AI의 공감: 정서적 이해를 위한 대규모 언어 모델 평가

Empatia em IA: avaliando grandes modelos de linguagem para compreensão emocional

Эмпатия в ИИ: оценка больших языковых моделей для эмоционального понимания

Yapay Zekada Empati: Duygusal Anlayış İçin Geniş Dil Modellerinin Değerlendirilmesi

Sự đồng cảm trong AI: Đánh giá các mô hình ngôn ngữ lớn để hiểu cảm xúc

人工智能中的同理心：评估大型语言模型以实现情感理解

Stories, News & Mentions about OPUS

Opus’s Stories on HackerNoon

Opus's News & Mentions