Andrew McAfee

Lin Guan and Yifan Zhou have led a very interesting study investigating the effectiveness of Vision Language Models in critiquing robot behaviors in terms of the desirability of their "style." If you are wondering how our work on LLM's inability to verify correctness squares with this study on using them as style critics, see my earlier post https://lnkd.in/gBYAr5qq ! This is the reason why being a style critic is one of the roles LLMs can play in LLM-Modulo framework: https://lnkd.in/gyAyKx4E Check out the details of the work @ https://lnkd.in/gfVZ4EYW ( Lin Guan also has a nice twitter thread on the work here: https://lnkd.in/gnieU-qM )

89

2 Comments

Essential #AI thoughts from Fei-Fei Li. I was honored to hear her speak about AI years ago at orientation for new fellows at The Paul & Daisy Soros Fellowships for New Americans and I never forgot her wisdom and prescience. She heads up HAI (Human-Centered AI at Stanford University. Q: What do you you think about AI’s future, what worries you the most, and what gives you hope? A: It’s not AI’s future, it’s humanity’s future. We don’t talk about electricity’s future, we don’t talk about steam’s future. At the end of the day, it is our future, our species’ future, and our civilization’s future—in the context of AI. #futureofeducation #futureofwork

28

Quantum computing is here and growing at 32% per year growing to a $6.5B market by 2030. Great overview from the Massachusetts Institute of Technology on what as a leader you need to know. This will be like throwing gasoline on a fire when it comes to processing power for artificial intelligence, machine learning, and data analytics across large data sets. You need to have a plan now, especially from a cyber security perspective to counter AI & quantum enabled attacks. Take the time to think about how quantum will fit into your short and long term strategies now to take advantage of what it will bring in your market as well. Will your applications be able to take advantage of this compute power? How will you access quantum capability to differentiate in the market with your own data? Let us know if we can help in any way with our global team and network of trusted partners at Red Summit Board Services & Advisory Take care of yourself, and if you can, someone else!

7

OpenAI is undergoing significant restructuring and internal changes related to status and sparking an exodus of high-profile executives. #datascience #AI #artificialintelligence https://hubs.li/Q02ShsZy0 If you enjoyed this post, follow me for more insights on AI and data science!

13

1 Comment

What a difference a dozen years make! In 2012, we created and launched the White House Presidential Innovation Fellows (PIF) program. We could barely imagine at that time the extent and rapidity of the progress that was about to occur in machine learning and related fields. But we believed deeply that having a pathway for getting technologists into high-impact public service mattered. Yesterday, as part of President Biden's AI Talent Surge, the US Government announced a new PIF cohort consisting exclusively of leading AI experts to help ensure that eight key departments -- Justice, State, OMB, DoD, and more -- responsibly develop and use AI tools in order to effectively achieve their missions on behalf of the American people. Just as every organization became a tech organization in recent decades, so too will they need to graft AI onto their DNA in order to meet the rising expectations of the public. I'm heartened that these AI PIFs answered the call and will now use their diverse backgrounds and invaluable experiences in data, systems design, governance, equity, and security to make sure that AI works for everyone. Learn more here: https://lnkd.in/ebwXZxmi #AI #ML #DigitalGov #TechForGood #USA

132

2 Comments

I had the privilege of sitting down with Ida Momennejad, whose years of brilliant research on multi-step planning and cognition could not be more timely and important amid today's conversations around agentic AI. Here are some of the highlights from our conversation, many from her published papers. I encourage you to read more and share your thoughts, linked below. - Despite many claims on what LLMs can do, they're not good at executive functions like abstract reasoning and multi-step planning. Importantly, to understand where the gaps are in their ability (and to close these gaps), we need to have methods for proper evaluation established. Ida studied if LLMs could extract cognitive maps to use in planning and inference (spoiler alert - they could not). See: https://lnkd.in/gQSQXwQf - To enable agents and systems to perform better at these kinds of tasks, Ida proposes that LLMs need a prefrontal cortex and thus recommends a modular system approach for agentic AI. Different modules (agents) function as the generator, orchestrator, evaluator, predictor, etc. They work together in different roles to better accomplish tasks that require multi-step planning and accomplishing compositional tasks like decision making. This is similar to the framework also seen in the Microsoft's Autogen framework. See: https://lnkd.in/gfaRYMqA https://lnkd.in/gWFETWk7 - We wrapped the conversation on Human-AI alignment. Her recent paper on multi-player games studied the differences in behavior in humans vs. AI agents in game play. This is important in scaling AI safely, as well as better designing behavior in agents that humans enjoy interacting with. Ida proposes the “Task-sets” framework, an interpretable approach towards human-AI alignment. See: https://lnkd.in/gXz4ti56 #AI #TechWeek #GenAI #agents #AIAlignment #reinforcementlearning

50

4 Comments

Kudos to the Massachusetts Institute of Technology for convening one of the largest interdisciplinary gatherings of #AI leaders, entrepreneurs, researchers and academics. Thank you to John Werner, Sandy Pentland and Randall Lane for bringing us together. Together with fellow panelists Jeffrey Saviano, Dr. Sasha Luccioni and Ra'ad Siraj we excavated AI Governance, Sustainability, and Scale, and the imperatives of ethics, corporate values and responsible business with a principled-based approach to #AI. As with driving, like business, brakes let you go faster. It feels counter-intuitive, but without brakes we would drive really slowly. A great moment for advancing #AI, with the MIT Media Lab hosting 2000+ attendees, 180 speakers, 80 AI startups, 67 "lightning talks", 36 panels, and 35 technical talks, spread over 4 tracks. #artificalintelligence #startups #entrepreneurship #mit #tech #esg #sustainability #LLM #generativeAI

140

5 Comments

This is just incredible. Neal Ghosh's demonstration of how #VentureIQ takes deeply technical and complicated content across a range of industrial use cases and ages/stages through an AI-generated conversational podcast (a la #notebookLM) is really mind-blowing. Check out his post, below, and if you haven't seen VentureIQ, yet, let's get you set up to learn more. You can visit VentureIQ.ai or DM any of us over here to book a demo!

2

From Dr. Jeffrey Funk. Most entrepreneurs are not, in general, technologists. Some are, but the overwhelming majority of people who go into business out of an MBA program are usually far more interested in financial returns than in the technology itself. This often means that there is usually a huge amount of magical thinking that takes place in business, because these are NOT experts in the technology, and they are dependent not only upon their pet wizards (who usually are only secondarily interested in the monetization aspect) but also on their own understanding and perception about the technologies and the markets these technologies address. When such technology gets beyond the skill sets even of most "experts" in that space (which is where we are now), the ability to evaluate that technology and its potential becomes even more suspect. While I think it is still "early" in terms of the potential of generative AI in particular, there are several key holes that still need to be filled first before it becomes more than a moderately break-even proposition. These aren't being taught in business schools, and there's real danger there because of it.

10

Learn AI with Anand's highly sought-after Live Cohort, "Mastering Transformer Models for GenAI LLMs," for the August 2024 cohort, is filling up quickly. Enrollment closes today at midnight (IST), with only a few spots left. Provisional Registration: https://lnkd.in/gA4NRKPH #ai #artificialintelligence #genai #greatlearning #VIT #dataanalyst #datascientist

29

Eric Topol, MD interviews Kate Crawford, professor at both USC and Microsoft Research and leading scholar of the social implications of #AI as they attempt to answer the question: What are the costs #generativeAI is incurring on the planet, and also on us as a people? Some prominent ethical questions the interview raises that I found particularly disconcerting: 🙋‍♀️ What sorts of biases are hidden in AI models and how will they be propagated if we start to use AI for healthcare, education or even criminal justice decisions? 🙋‍♀️ Who are the humans doing the hard work of labelling material while being paid less than $2/hour to generate a training model? 🙋‍♀️ How can we create a system that will continue to reward artists, writers and musicians for the hard work they do when some of their work can also be ingested by AI and reproduced in a way that is not copyrighted? #AIethics is not just about how #generativeAI is being used and how to create policy to govern its use, but also what the hidden costs are of continuing to invest in it as a technology 🌎 I'm encouraged to know that researchers like Kate Crawford are looking into how these systems are being built, who might be harmed and how we can mitigate the risks that come with this big shift towards #AI. #AIethics #ethicsinTech #AIRegulation https://lnkd.in/e_62ZetE

8

1 Comment

The keynotes for #ODSCWest 2024 represent the best in the field of AI, featuring experts from UC Berkeley, NVIDIA, Anthropic, Voltron Data, MIT, and Google Cloud. #datascience #AI #artificialintelligence https://hubs.li/Q02ShsR50 If you enjoyed this post, follow me for more insights on AI and data science!

4

We are here in DC talking AI - looking forward to an insightful conversation. Panel Title: AI the Scholarly Ally: How is Gen AI both Facilitating and Undermining Scientific Publishing? Last one today!

3

One of the reasons I am bullish on cognitive/behavioral scientists in the age of Gen AI is that they think deeply about human-AI computer interaction. Some people want complete automation. Others want a hybrid - use AI for the annoying, repetitive tasks, and leverage human expertise for the most important aspects. This idea was a core part of my PhD research. Humans and AI can collaborate in scientific theory development, building better theories in shorter periods of time. The AI performed well at finding models that predicted the data well. Humans were skilled at dissecting these uninterpretable models and being able to *explain* them. How can AI and humans collaborate in this new age of Gen AI? We are a little obsessive about this at Roundtable if you can't tell...

24

2 Comments

My epic co-founder Tim Cheung did an excellent job (as always) in his latest post for Factory AI. The article is about how manufacturers can leverage the goldmine they’re sitting on—their data. First, he clears some of the fog on today’s hot topics: AI, ML, Deep Learning, and Gen AI. Second, he shares six relevant use cases, each accompanied by a case studies for each. Check out the link to the post in the comments. Give it a read and let us know your thoughts as always. #Manufacturing #FoodAndBeverage

16

3 Comments

Adding to the announcement from Kapor Center, Omidyar Network and San Francisco Foundation check out these purposes--so vitally important to take this integrated and interdisciplinary approach: Kapor Foundation, Omidyar Network, and San Francisco Foundation believe that California is uniquely positioned to lead the nation in setting the standards of what responsible, equitable and ethical AI looks like and ensuring all Californians can benefit from its promise and potential. This will require investments across public, private, and nonprofit sectors to: 1) Invest in world-class education to ensure all students are prepared to be critical consumers and producers of emerging technologies 2) Support new approaches to talent development and worker power to prepare the underrepresented workforce for jobs in the emerging technology sector Invest in the development of new technologies and products to address critical inequities across sectors while producing economic gains 3) Advocate to put appropriate regulation and accountability measures in place to protect from potential harms while spurring innovation 4) Investing in innovative solutions to solve the Bay Area's housing crisis so that everyone can have affordable and secure housing Thoughtful and pathbreaking! #ai #foundations #equity #gapclosing #dei #technology

2