URL and discussions of the paper "Can AI agents conduct open-ended AI research? Early evidence from two case studies"
📢 New paper: Forecasts of explosive AI progress hinge on AI agents automating AI research. But most evaluations of agents conducting AI research focus on narrow, verifiable tasks. Can AI agents… | Sayash Kapoor
📢 New paper: Forecasts of explosive AI progress hinge on AI agents automating AI research. But most evaluations of agents conducting AI research focus on narrow, verifiable tasks. Can AI agents conduct open-ended research? https://lnkd.in/gfP-q4CD We gave agents research questions from two unpublished papers, six days, and thousands of dollars of API credits and compute. The authors of the original papers reviewed the AI-generated papers. They unambiguously rejected agents' outputs. Agents were fluent at most *engineering* tasks. They conducted serious literature reviews, debugged GPU environments, ran hundreds of experiments, and turned in camera-ready LaTeX without human help. We also found no evidence of reward hacking. If anything, we found the opposite: the agents started with marketable claims and walked them back to negative results as the evidence came in. But neither agent output was close to the bar of a top conference paper. Both papers suffered from similar failures: poor judgment about the bar for an AI paper submitted to a top conference, the lack of creative problem solving and ineffective backtracking, poor awareness of resources, and instruction drift. This research design has many limitations: the small sample size, non-blind reviews, and the reviewers knowing that the work was AI-generated. We also couldn't test Anthropic's strongest model, because Fable 5 is deliberately limited on frontier AI research tasks, so ended up using OpenClaw with Opus 4.8 (extra-high) for our main experiments and Codex with Sol 5.6 (ultra) for a robustness check. But we think the research design is still helpful in assessing AI agents' ability to conduct research, and it is complementary to evaluations on verifiable tasks, as well as blinded reviews of AI outputs. In follow-up studies, we are expanding the set of non-public papers we evaluate. If you are an AI researcher with unpublished papers, we would love to collaborate with you on our next evaluation. Expression of interest: https://lnkd.in/gpeykJea We also release the agent logs and all the code and data, so that others can conduct their own analyses of our results: https://lnkd.in/gJarPAnb Finally, we plan to conduct such evaluations regularly, and are hiring a senior researcher to help lead these efforts. Apply here: https://lnkd.in/erJZdmve I'm grateful for the core team leading this effort: Peter Kirgis, Andrew Schwartz, Stephan Rabanser, and Arvind Narayanan, and to our collaborators who reviewed AI papers, analyzed agents logs, and gave feedback on the paper: David Demitri Africa, Konstantinos V., Viet Nguyen, Dr Toby D. Pilditch, Magda Dubois, Harry Coppock, Cozmin Ududec, Nitya Nadgir, Matilda Orona, Tilman Bayer, Derrick Chan-Sew, Eric (Yue) Ling, Abhishek Shetty, Helen Toner, Gillian K. Hadfield, Seth Lazar, Steve Newman, Shoshannah Tekofsky, Rishi Bommasani
Science that Compounds: The Need for A New Substrate for Research in the Age of AI
This paper is a perspective from Lightcone Research, an open-source initiative building tooling for scientific research in the age of agentic AI.
No. 020: Astra solves ten open math problems, AI agents fail at open-ended research, peer review under strain - The Hallway Track
The Hallway Track No. 020 - How AI is affecting the practice of science
How to build an AI scientist: first peer-reviewed paper spills the secrets
AI Scientist, an autonomous research tool, first released in 2024, has now undergone peer review, highlighting its strengths and limitations.
Can AI agents conduct open-ended AI research? Early evidence from two case studies
Forecasts of explosive AI progress hinge on AI agents automating AI research. But evidence on whether agents can carry out open-ended AI research is thin. Current evaluations either test agents on narrow, verifiable tasks, which excludes open-ended research, or submit AI-generated papers to blind peer review, which is overstretched, stochastic, and suffers from poor review quality. We introduce a third way to measure progress towards AI R\&D automation. An agent takes on the central, open-ended research question of a high-quality unpublished paper, and the paper's original authors grade its output. We call these shadow evaluations. We ran shadow evaluations on two unpublished NeurIPS 2026 submissions, giving frontier agents six days and thousands of dollars of compute. The agents completed all of the engineering without human help, yet could not make substantial progress towards answering the research questions. As a result, both papers were unambiguously rejected by the authors. We identify five recurring failure modes: poor judgment about the bar for publishable research, uncreative responses to shortcomings in the research design, ineffective backtracking from dead ends, poor resource awareness, and instruction drift. A robustness check with a second model and scaffold reproduced these failures. We release the expert reviews, survey responses, agent repositories, and logs. Our results provide early evidence that today's agents can do the engineering of AI research, but struggle with critical parts of the research lifecycle.
White Paper on Public AI – Open Future
Today's AI landscape is dominated by a few tech companies. This White Paper presents an alternative: Public AI systems built on transparency, democratic governance, and open access to critical infrastructure.
Will we have a theory for in-the-wild AI agents? - undisciplinary
Thoughts after writing a sociology-meets-physics-meets-agents paper I just put out a preprint out on a theory of role emergence in multi-agent systems (soci…
The argument against AI agents and unnecessary automation
Opinion: OpenAI's Operator a solution in search of a problem
Research acceleration: The view inside OpenAI
Inside OpenAI, coding agents are reshaping AI research. Explore early data on agent usage, experiment velocity, task complexity, and research acceleration.
Human scientists trounce the best AI agents on complex tasks
State-of-the-industry report finds that, despite the limitations of artificial-intelligence systems, researchers have embraced them.
Where will AI be at the end of 2027? A bet
We, Gary Marcus, author, scientist, and noted skeptic of generative AI, and Miles Brundage, an independent AI policy researcher who recently left OpenAI and is bullish on AI progress, have agreed to the following bet, at 10:1 odds, with criteria drawn from two earlier Substack essays by Gary that proposed criteria for
The Arguments Against Open Source AI are Very Bad | Tom Bedor's Blog
The release of Kimi K3 has opened a fresh round of angst and confused discourse. There's a loud cohort of journalists, business leaders, and politicians arguing that open source AI is a dangerous threat. OpenAI's Dean Ball:
Agents
Intelligent agents are considered by many to be the ultimate goal of AI. The classic book by Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach (Prentice Hall, 1995), defines the field of AI research as “the study and design of rational agents.”
For Most of the World, Open-Source AI Is the Only Way Forward
Proprietary AI is both too expensive and too centralized in control for most countries and companies to rely upon.
