Martin Soto

Featured in: Favicon

securityboulevard.com Favicon

devops.com Favicon

viatravelers.com

Articles

Tell me about yourself: LLMs are aware of their implicit behaviors — LessWrong

Jan 21, 2025 | lesswrong.com | Martin Soto |Martín Soto

This is the abstract and introduction of our new paper, with some discussion of implications for AI Safety at the end. Authors: Jan Betley*, Xuchan Bao*, Martín Soto*, Anna Sztyber-Betley, James Chua, Owain Evans (*Equal Contribution). We study behavioral self-awareness — an LLM's ability to articulate its behaviors without requiring in-context examples. We finetune LLMs on datasets that exhibit particular behaviors, such as (a) making high-risk economic decisions, and (b) outputting insecure code.
The Information: OpenAI shows 'Strawberry' to feds, races to launch it — LessWrong

Aug 27, 2024 | lesswrong.com | Martin Soto |Martín Soto

Two new The Information articles with insider information on OpenAI's next models and moves. They are paywalled, but here are the new bits of information:Strawberry is more expensive and slow at inference time, but can solve complex problems on the first try without hallucinations.
The need for multi-agent experiments — LessWrong

Aug 1, 2024 | lesswrong.com | Martin Soto |Martín Soto

TL;DR: Let’s start iterating on experiments that approximate real, society-scale multi-AI deploymentEpistemic status: These ideas seem like my most prominent delta with the average AI Safety researcher, have stood the test of time, and are shared by others I intellectually respect. Please attack them fiercely!Some authors have already written about multi-polar AI failure. I especially like how Andrew Critch has tried to sketch concrete stories for it.
OpenAI releases GPT-4o, natively interfacing with text, voice and vision — LessWrong

May 13, 2024 | lesswrong.com | Martin Soto |Martín Soto

This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
Conflict in Posthuman Literature — LessWrong

Apr 6, 2024 | lesswrong.com | Martin Soto |Martín Soto

Grant Snider created this comic (which became a meme):Richard Ngo extended it into posthuman/transhumanist literature:That's cool, but I'd have gone for different categories myself. Here they are together with their explanations. Top: Man vs Agency(Other names: Superintelligence, Singularity, Self-improving technology, Embodied consequentialism.)Because Nature creates Society creates Technology creates Agency.
Measuring Coherence of Policies in Toy Environments — LessWrong

Mar 18, 2024 | lesswrong.com | Richard Ngo |Martin Soto |Martín Soto

This post was produced as part of the Astra Fellowship under the Winter 2024 Cohort, mentored by Richard Ngo. Thanks to Martín Soto, Jeremy Gillien, Daniel Kokotajlo, and Lukas Berglund for feedback. SummaryDiscussions around the likelihood and threat models of AI existential risk (x-risk) often hinge on some informal concept of a “coherent”, goal-directed AGI in the future maximizing some utility function unaligned with human values.
Consciousness as intrinsically valued internal experience — LessWrong

Jul 10, 2023 | lesswrong.com | Matt Goldenberg |Matthew Barnett |Martin Soto |Martín Soto |Jonathan Erhardt

I'm afraid I'm sceptical that you methodology licenses the conclusions you draw. Thanks for raising this. It's one of the reasons I spelled out my methodology, to the extent that I had one. You're right that, as I said, my methodology explicitly asks people to pay attention to the internal structure of what they were experiencing in themselves and calling consciousness, and to describe it on a process level.