Our conclusion: "LLMs’ apparent ToM abilities may be fundamentally different from humans' and might not extend to complex interactive tasks like planning.". Preprint: Code: Demo: /end 🧵.

This work began at @DivIntelligence and is in collaboration w/ @nedcpr, @RasmusOvermark, Beba Cibralic, @nickhaber, and @camrobjones.

I'll be talking about this in SF at #CogSci2025 this Friday at 4pm. I'll also be presenting it at the PragLM workshop at COLM in Montreal this October.

LLMs are already deployed as educators, therapists, and companions. In our discrete-game variant (HIDDEN condition), o1-preview jumped to 80% success when forced to choose between asking vs telling. The capability exists, but the instinct to understand before persuading doesn't.

These findings suggest distinct ToM capabilities:. * Spectatorial ToM: Observing and predicting mental states. * Planning ToM: Actively intervening to change mental states through interaction. Current LLMs excel at the first but fail at the second.

Why do LLMs fail in the HIDDEN condition? They don't ask the right questions. Human participants appeal to the target's mental states ~40% of the time ("What do you know?" "What do you want?") LLMs? At most 23%. They start disclosing info without interacting with the target.

Key findings:. In REVEALED condition (mental states given to persuader): Humans: 22% success ❌ o1-preview: 78% success ✅. In HIDDEN condition (persuader must infer mental states): Humans: 29% success ✅ o1-preview: 18% success ❌. Complete reversal!

Setup: You must convince someone* to choose your preferred proposal among 3 options. But, they have less information and different preferences than you. To win, you must figure out what they know, what they want, and strategically reveal the right info to persuade them. *a bot

I'm excited to share work to appear at @COLM_conf! Theory of Mind (ToM) lets us understand others' mental states. Can LLMs go beyond predicting mental states to changing them? We introduce MINDGAMES to test Planning ToM--the ability to intervene on others' beliefs & persuade them.

RT @harveyiyun: LLMs excel at finding surprising “needles” in very long documents, but can they detect when information is conspicuously mi….

This is work done with. @declangrabbmd.@kevin_klyman.@willie_agnew.@snchancellor.@nickhaber.@_desmond_ong . Thanks ❤️.

📝Read our pre-print on why "Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers" here:.

📋We further identify **fundamental** reasons not to use LLMs as therapists, e.g., therapy involves a human relationship: LLMs cannot fully allow a client to practice what it means to be in a human relationship. (LLMs also can't provide in person therapy, such as OCD exposures.).

🔎We came up with these experiments by conducting a mapping review of what constitutes good therapy, and identify **practical** reasons that LLM-powered therapy chatbots fail (e.g. they express stigma and respond inappropriately.

📈Bigger and newer LLMs exhibit similar amounts of stigma as smaller and older LLMs do toward different mental health conditions.

📉Large language models (LLMs) in general struggle to respond appropriately to questions about delusions, suicidal ideation, and OCD and perform significantly worse than N=16 human therapists.

🚨Commercial therapy bots make dangerous responses to prompts that indicate crisis, as well as other inappropriate responses. (The @APA has been trying to regulate these bots.)

🧵I'm thrilled to announce that I'll be going to @FAccTConference this June to present timely work on why current LLMs cannot safely **replace** therapists. We find. ⤵️

Still looking for a good gift?🎁 Try my book, which just had its first birthday! @KirkusReviews called it a "thought-provoking tech tale.”. @kentarotoyama said it "reads less like sci-fi satire and more as poignant, pointed commentary on homo sapiens"

I just landed in Miami to present at @emnlpmeeting the work I did with @Diyi_Yang from @stanfordnlp. Please reach out if you'd like to meet!. And read @StanfordHAI's post about our work here:.