Reinforcement Learning News

6. Many more AI companies will begin building custom chips.

Dec 28, 11:29 AMDec 31, 11:59 PM

90.5%chance

246554

OptionVotes

YES

3092

324

Will reinforcement learning overtake LMs on math before 2028?

May 29, 11:10 PMJan 2, 4:59 AM

60.78%chance

394345

OptionVotes

YES

1390

541

Will the first AI to get IMO gold cite AlphaZero?

Dec 11, 6:03 AMNov 11, 11:59 PM

54.82%chance

191736

OptionVotes

YES

1251

458

Will an autonomous robot beat an Olympic champion at table tennis before 2028?

Apr 22, 7:20 PMDec 31, 8:08 PM

21.57%chance

121502

OptionVotes

YES

1907

524

Will Emmett Shear's new company make important and central contributions to AI alignment?

Mar 29, 8:45 PMDec 31, 11:59 PM

20.14%chance

141487

OptionVotes

YES

1991

502

Will we get rational reinforcement learning before 2027?

May 11, 6:19 AMDec 30, 11:59 PM

35.55%chance

121107

OptionVotes

YES

1349

746

Reinforcement Learning with Human Feedback (permanent)

Mar 27, 2:23 PMAug 11, 9:43 PM

19816

OptionProbability

YES

368

300

Inverse Reinforcement Learning (IRL) a significant component of publicly known SOTA alignment techniques by 2028

Sep 16, 9:35 AMJan 1, 7:59 AM

14.13%chance

15490

OptionVotes

YES

1236

906

If Tetra draws a shoggoth about this paper or post, will it get >3K views in 24 hours?

Oct 7, 10:00 PMJan 1, 10:49 PM

380

OptionProbability

FutureLivingLab, iFlow ROME, Safety-Aligned Data Composition https://huggingface.co/papers/2512.24873

MIRI, "How could an LLM destroy humanity?" short https://youtube.com/shorts/Z6oPeFvp6_s

MIRI, An International Agreement to Prevent the Premature Creation of Artificial Superintelligence, Notes on Article VI https://arxiv.org/abs/2511.10783

Epoch AI, How many AI agents could run on the AI chips shipped through 2027? https://epoch.ai/publications/estimating-the-agent-population

Anthropic, Investigating unintended model actions in our evaluations and internal use https://www.anthropic.com/news/investigating-unintended-model-actions

Kirgis & Hua, The Continual Constitutional Convention: How language models revise the documents that govern them https://www.alignmentforum.org/posts/c9woewdY9ddsPqf3J/the-continual-constitutional-convention-how-language-models

Anthropic, Rosu & Wang, Training Alignment Auditors via Reinforcement Learning https://alignment.anthropic.com/2026/training-alignment-auditors/

100

Group relative rewards in RL (e.g. GRPO) teaches AIs to be spiteful, in publication by EOY 2027.

Sep 27, 7:31 AMDec 31, 11:59 PM

62.17%chance

230

OptionVotes

YES

128

Will a paper titled "maximum sparsity reinforcement learning" come out by EOY2027?

Jan 2, 9:15 PMJan 1, 7:59 AM

50%chance

110

OptionVotes

YES

100

Will Polychromic Objectives for Reinforcement Learning be generalized to manifolds by 2050?

Jul 1, 8:06 PM

Reinforcement Learning News

Prediction markets for Reinforcement Learning

Prediction markets for Reinforcement Learning

Prediction markets for Reinforcement Learning

6. Many more AI companies will begin building custom chips.

Will reinforcement learning overtake LMs on math before 2028?

Will the first AI to get IMO gold cite AlphaZero?

Will an autonomous robot beat an Olympic champion at table tennis before 2028?

Will Emmett Shear's new company make important and central contributions to AI alignment?

Will we get rational reinforcement learning before 2027?

Reinforcement Learning with Human Feedback (permanent)

Inverse Reinforcement Learning (IRL) a significant component of publicly known SOTA alignment techniques by 2028

If Tetra draws a shoggoth about this paper or post, will it get >3K views in 24 hours?

Group relative rewards in RL (e.g. GRPO) teaches AIs to be spiteful, in publication by EOY 2027.

Will a paper titled "maximum sparsity reinforcement learning" come out by EOY2027?

Will Polychromic Objectives for Reinforcement Learning be generalized to *manifolds* by 2050?

Latest stories

Will Polychromic Objectives for Reinforcement Learning be generalized to manifolds by 2050?