Does automation look more like a crashing wave hitting some workers and not others, or a broadly rising tide? Our latest Substack post from @adamkuzee explains the concepts and data behind a recent MIT FutureTech paper on the shape of AI progress: mitfuturetech.substack.com/p… @MITSloan @MIT_CSAIL @ProfNeilT #AI #Automation #FutureOfAI #FutureOfWork #AIResearch #MITFutureTech

@porat_ruth leads the discussion on global #AIAdoption with @elerianm and FutureTech Research Scientist @Fleming_Martin at the #google #mitfuturetech #AIfortheEconomy Forum Read our latest research on the economics of human AI collaboration here: arxiv.org/html/2603.29121v1

@ProfNeilT kicks off the #google #MITFuturetech #AIfortheEconomyForum with our findings on AI automation from Thousands of Worker Evaluations of Labor Market Tasks. Read the full paper: arxiv.org/abs/2604.01363

Can academic science keep up with the AI frontier? Our substack tracks the meteoric rise of AI foundation models in science and the constraints on future breakthroughs: mitfuturetech.substack.com/p… #AI #Science #AIResearch #FutureOfScience

MIT FutureTech is hiring a Research Assistant to work with Dr. @DanialLashkari on projects at the intersection of technological progress, innovation, and AI Join us! futuretech.mit.edu/opportuni… #EconRA #AIEconomics #ResearchJobs #AIResearch

Extending our gratitude to Nobel Prize Laureate and Turing Award Winner @geoffreyhinton for speaking at the MIT FutureTech Lab Seminar, we were honored to have him speak. Read the full presentation: "Living with Alien Beings" drive.google.com/file/d/1taG… #AISafety #AIResearch

Senior Data Scientist Anna Li points out the limitations in quantum computing as a solution to the AI memory wall. “We would need to reinvent the entire stack, which would take a long time to mature.” #AIMemoryWall #quantumComputing #AIConference #MITFutureTech

That's a wrap on Automation in the Labor Market and Economy: AI Automation with Generative World Models! Thank you @du_yilun #GenerativeAI #AIResearch #MITFutureTech #AIConference

A great reality check from @beena_ammanath - Success in AI deployment in an enterprise setting is driven by adoption - when accurate models are rolled out but not adopted, it's considered a failed project! #AIAdoption #EnterpriseAI #AIDeployment #MITFutureTech

Congratulations to @ProfDavidDeming on his new position: Danoff Dean of Harvard College @harvard! Fantastic presentation today at #MITFutureTech Conference "The Rapid Adoption of Generative AI", read it here: nber.org/papers/w32966

Does change in expertise (caused by task automation) affect change in wages? @davidautor says YES! Read the full paper on expertise here: nber.org/papers/w33941 📷#TaskAutomation #EconomicsofAutomation #AIEconomics #LabourResearch

#MITFutureTech Conference day 1 drew to a close with an inspirational discussion on Modern Challenges in Automating science led by @ta_broderick #AIResearch #AIAutomation #AutomationinScience

#MITFutureTech Conference day 1 drew to a close with an inspirational discussion on Modern Challenges in Automating science led by @ta_broderick #AIResearch #AIAutomation #AutomationinScience

Patrick Rich from @IBMResearch shares how foundation models accelerate scientific discovery: from experiments to algorithms. “We are at the dawn of a new algorithmic era powered by new representations of information” #foundationmodels #AIResearch #MITFutureTech

AI Adoption in Biology is accelerating new approaches to Drug Discovery via methods such as Cell Painting, as discussed by @shantanuXsingh from The Broad Institute. Advances in technology don't necessarily result in decreased costs in drug discovery however, a phenenom dubbed "Eroom's Law" - the opposite of Moore's law.

Excellent perspective from @KarianneBergen on educating the next generation of researchers in the context of #GenAI #AIConference #AIResearch #MITFutureTech

@GaryMarcus delivers a bitter lesson on the bitter lesson - scaling works well only for some problems - mainly pattern recognition - but not others #AIExpert #Scaling #MITFutureTech #AIConference

Can the entire scientific process be automated? Research Scientist Yutaro Yamada explores AI Scientists and Self Improving AI #AIScientist #AIResearch #MITFutureTech

Are you a technology optimist or a technology pessimist? Speaker Kristina McElheran says Yes! Read her paper covering AI Adoption in America: nber.org/papers/w31788 #AIAdoption #AIResearch #AIConference #MITCSAIL

Lab Director Dr. Neil Thompson kicks off the conference with key FutureTechs Research, get an overview on our new substack! substack.com/@mitfuturetech #AIResearch #AIEconomics

Welcome to day 1 of the FutureTech Conference: How Fast Will AI Automation Happen? Follow along for live updates! #AIConference

In this @TEDTalks event, Neil Thompson (@ProfNeilT) discusses the quantification of the algorithmic efficiency improvements happening in deep learning and how these have reduced the environmental and energy demands of large language models. piped.video/naDR7siJMqE?si=89UE…

Our director, Dr. Neil Thompson (@ProfNeilT), recently had his National Bureau of Economic Research working paper with David Autor (@davidautor) featured in the Financial Times (@FinancialTimes). Access the article here: on.ft.com/4lxAXS4 Read the working paper here: nber.org/papers/w33941

A new paper from David Autor (@davidautor), in collaboration with Neil Thompson (@ProfNeilT), makes an important contribution to explaining how AI is likely to impact labor markets. Based on a rigorous model, confirmed with an analysis of 40 years of data, they provide a nuanced perspective on how automation impacts job employment and wages. Essentially, this depends on the extent to which easy tasks are removed from a role and expert ones are added, and how specialized a role becomes as a result. When jobs gain inexpert tasks but lose expertise, wages decline, but employment may increase. Think of how taxi driving became less specialized, and well-paid, but more common, due to Uber. In contrast, when technology automates the easy tasks inside a job, the remaining work becomes more specialized. Employment falls because fewer people now qualify, but the scarcity of expertise drives wages up. This is what seems to be happening with proofreading, which is now less about spell-checking and more about helping people to write, leading to lower job numbers but higher average wages. Their model helps us to understand the impacts of AI on labor markets. For instance, why AI tools can raise wages for senior software engineers, but decrease employment, while simultaneously reducing earnings, and increasing employment, for more entry level software engineering roles. Read: nber.org/papers/w33941 See also this talk from David at Stanford HAI (@StanfordHAI): piped.video/watch?v=uV3Lttj8…

Peter Slattery (@PeterSlattery1) recently coauthored this paper: "We directly compare the persuasion capabilities of a frontier large language model (LLM; Claude Sonnet 3.5) against incentivized human persuaders in an interactive, real-time conversational quiz setting. In this preregistered, large-scale incentivized experiment, participants (quiz takers) completed an online quiz where persuaders (either humans or LLMs) attempted to persuade quiz takers toward correct or incorrect answers. We find that LLM persuaders achieved significantly higher compliance with their directional persuasion attempts than incentivized human persuaders, demonstrating superior persuasive capabilities in both truthful (toward correct answers) and deceptive (toward incorrect answers) contexts. We also find that LLM persuaders significantly increased quiz takers’ accuracy, leading to higher earnings, when steering quiz takers toward correct answers, and significantly decreased their accuracy, leading to lower earnings, when steering them toward incorrect answers. Overall, our findings suggest that AI’s persuasion capabilities already exceed those of humans that have real-money bonuses tied to performance. Our findings of increasingly capable AI persuaders thus underscore the urgency of emerging alignment and governance frameworks. " "the infrastructure, practices, and norms for reporting flaws in general-purpose AI (GPAI) systems remain seriously underdeveloped, lagging far behind more established fields like software security. Based on a collaboration between experts from the fields of software security, machine learning, law, social science, and policy, we identify key gaps in the evaluation and reporting of flaws in GPAI systems. We call for three interventions to advance system safety. First, we propose using standardized AI flaw reports and rules of engagement for researchers in order to ease the process of submitting, reproducing, and triaging flaws in GPAI systems. Second, we propose GPAI system providers adopt broadly-scoped flaw disclosure programs, borrowing from bug bounties, with legal safe harbors to protect researchers. Third, we advocate for the development of improved infrastructure to coordinate distribution of flaw reports across the many stakeholders who may be impacted. These interventions are increasingly urgent, as evidenced by the prevalence of jailbreaks and other flaws that can transfer across different providers’ GPAI systems." arxiv.org/pdf/2505.09662

We were cited in this report by the Bipartisan House Task Force on Artificial Intelligence. republicans-science.house.go…

Peter Slattery (@PeterSlattery1) recently spoke about his work on the MIT AI Risk Repository (@MITAIRisk) on a panel about the future of AI at the U.S. Securities and Exchange Commission (@SECGov) in Washington, DC. See agenda here: sec.gov/newsroom/press-relea…

Resharing our Quantum Economic Advantage Calculator on #WorldQuantumDay futuretech.mit.edu/quantum-e… Great work from Neil Thompson, Jayson Lynch, and Johannes Galatsanos.

At a recent lab meeting, Ted Xiao from Google DeepMind discussed progress in Robot Foundation Models. This relates to work which Sebastian Sartor and Neil Thompson have been leading (see link in comments).

Jayson Lynch recently collaborated on this article and benchmark with Clinton Wang, Dean Lee, Cristina Menghini, Johannes Mols, Jack Doughty, Adam Khoja, Sean Hendryx, Summer Yue, and Dan Hendrycks.

Christophe Combemale with his final closing remarks, including the importance of widespread inclusion in the decision-making around how AI should affect our collective future.

Martin Fleming (@Fleming_Martin) with some closing remarks, including the importance of ensuring that we don't automate a broken system.

Susan Helper, a Professor from the Weatherhead School of Management at Case Western Reserve University, sharing her reflections and learnings from the workshop.

A panel on 'Data to Support Research on Economics of AI' chaired by Christophe Combemale, and featuring Morgan Frank (@mrfrank5790), Erica Groshen, and Jason Owen-Smith.

Jason Owen-Smith, a Professor from the University of Michigan, discussing a decentralized system for collecting data relevant to AI's employment impacts.

Erica Groshen, a Senior Economic Advisor from Cornell University, discussing systems to curate and integrate data relevant to AI's impact on employment.

Morgan Frank (@mrfrank5790), Assistant Professor in the Department of Economics at the University of Pittsburgh (@PittTweet), discussing his work to develop datasets, and models, that can help us to predict AI unemployment risks.

A panel on 'Industrial Organization and Structural Change' chaired by Morgan Frank (@mrfrank5790), and featuring Avinash (Avi) Collis (@avi_collis) and Youngjin Yoo.

Youngjin Yoo, Associate Dean of Research, Weatherhead School of Management at Case Western Reserve University, discussing AI diffusion.

Avinash (Avi) Collis (@avi_collis) from @HeinzCollege of Information Systems and Public Policy discussing GDP-B, an approach to assess the economic benefits of free goods, including generative AI.

Today is the second day of an event, AI: Who Wins and Who Loses, which we are running in partnership with the Carnegie Mellon University Block Center for Technology and Society (@CMUBlockCenter). See the agenda attached. We will be posting updates throughout the day.

A panel on 'Labor Supply and the AI Workforce', chaired by Erica Groshen, featuring Stuart Andreason (@StuartAndreason), and Adam Leonard.

Adam Leonard, a former Director of Information, Innovation & Insight, Texas Workforce Commission, discussing the importance of adapting education.

Stuart Andreason (@StuartAndreason), from The Burning Glass Institute (@TheBGInstitute), discussing the impacts of AI on labor supply, and how regional differences.

Erica Groshen, a Senior Economic Advisor from Cornell University, introduces Session 5, and discusses how we should respond to workforce impacts from AI.

A panel on 'Labor Demand Implications of AI', chaired by Christophe Combemale, featuring Jason Owen-Smith, Susan Helper and Rob Seamans (@robseamans).

Susan Helper, a Professor from Weatherhead School of Management at Case Western Reserve University (@cwru), discussing whether AI could improve job quality.

Jason Owen-Smith, a Professor from the University of Michigan, discussing their approach for collecting data to explain how AI is reshaping employment.