Fei

Effects are irreversible and therefore high-stakes agents shall be treated differently. Wrote the full-scope blogs and check it out getdonut.ai/blog/constrain… here

A simple note on AI in SF now, borrowed from Alan Turing, “I am afraid that the following syllogism may be used by some in the future: Turing believes machines can think, Turing lies with men, Therefore machines do not think. Yours in distress.”

For those who need context of the drama: It all starts with US reviewing Benchmark’s investment into Manus; and now the chapter has turned to China blocking Manus deal under the name of foreign investment reviews. Symmetrical reviews; then the climax?

@riskstate_ai @DonutAI Yes what could be parameterized would therefore be personalized

best piece i've read on agentic trading architecture this year. "if the model can see the rule, it can optimize around it" is the insight most builders miss — they put risk limits in the prompt and call it governance. the separation between planning and constraint layer is exactly right. the model proposes, a deterministic system adjudicates. that's the only way hard constraints stay hard. one thing i'd push further: the constraint layer itself should be dynamic. exposure caps that adapt to live volatility, macro regime, funding extremes — not just static user-set thresholds. the boundary needs to move with the market, not just hold a fixed line. been building exactly this separation. great to see the architecture formalized.

Constrained Autonomy Lessons from building D0, a proactive high-stakes agent for financial markets We built D0 for one thing: financial markets. This post is about a broader class of systems. getdonut.ai/blog/constrain…

For some reasons, I would love to read Quantum Suicide in Many-Worlds Implied P=NP (arxiv.org/abs/2603.28869) again against recent intelligence advances.

Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference @NeurIPSConf We have ethical reviews for authors, but missed it for invited speakers? 😡

Rosalind Picard made gross racist comments at NeurIPS 2024. we urge #mit and the #us Department of Education to investigate her and her claim. there is no place for rasicts in #ml and #ai. don't stop sharing this! help exposure! we need her investigated!

@RL_Conference @c_voelcker Additionally could we change the order of authorship in the final submission?

@c_voelcker Changes are totally fine, we just want the abstracts in to help us assess if we need more reviewer recruitment. It's not to constrain the final abstract

To accommodate nearby conferences and deadlines yet still maintain good reviewer matching, 𝐰𝐞 𝐚𝐫𝐞 𝐞𝐱𝐭𝐞𝐧𝐝𝐢𝐧𝐠 𝐭𝐡𝐞 𝐟𝐮𝐥𝐥 𝐩𝐚𝐩𝐞𝐫 𝐝𝐞𝐚𝐝𝐥𝐢𝐧𝐞 𝐭𝐨 𝐌𝐚𝐫𝐜𝐡 𝟖𝐭𝐡 but keeping March 1st as the abstract deadline