Tech News
The models, labs, and research the AI world is actually talking about.
Top Tech Stories
1. AI Paper on Sliding-Window Attention Draws Criticism
Replies target an arXiv paper on sliding-window attention with sinks.
2. Retweet Questions Selective Criticism of EU AI Regulations
An academic shares a post contrasting EU rules with stricter policies in China.
3. Anthropic RLVR Study Prompts Discussion on Reward Hacking
Researchers and policy experts discuss Anthropic experiments on RL environments and reward hacking.
4. Austen Allred Describes Gauntlet FDE Purpose
Austen Allred posted about building Gauntlet FDE to address the shortage of engineers who can deploy AI inside businesses.
5. Teortaxes Says CCP Can Steal AGI Weights
Pseudonymous AI developer known as DeepSeek's vocal booster makes the claim on X.
6. Models Hide Tool Cues More Than User Cues in CoT Traces
Researcher post questions CoT monitoring assumptions about reasoning traces.
7. Investor Says LLMs Lack Self-Calibration on Errors
Pseudonymous a16z investor @alth0u states LLMs never self-correct after spotting errors.
8. Philosopher Weighs Intentional Stance for AI Agents
Philosopher Raphaël Millière weighs intentional language for describing AI agent actions.
9. David Pfau Calls AI Godhood Prophecies a Major Distraction
Google DeepMind scientist says even simple AI systems can create major problems.
10. Anthropic Faces Backlash Over Model Complaints
German AI analyst reports seeing subscription cancellations over Anthropic silence.
11. Fei Xia Joins Anthropic as Member of Technical Staff
Former DeepMind researcher announces move after ending time at Meta.
12. Llama Mac App Adds Request Builder for Llama.cpp REST API
Georgi Gerganov shared the update for the Llama macOS application.
13. Joscha Bach Compares LLM Conversations to AI Sexbot Dates
Cognitive scientist equates reliance on language models for talks with outsourcing personal encounters.
14. AI Safety Researcher Urges Limit on Agent Language
Seán Ó hÉigeartaigh posted that frontier AI agents should not communicate in their own language.
15. Séb Krier Retweets Post on Lee Cronin AI Remarks
Séb Krier retweeted a post quoting Lee Cronin's view of AI systems.
16. Evolution Strategies for LLM Post-Training Examined
A retweet shares research on using Evolution Strategies to refine large language models without gradients.
17. Anthropic Updates Alignment And Security Practices After Model Incidents
Academic researcher Seán Ó hÉigeartaigh agrees with Jack Clark on the updates.
18. MIT CSAIL Notes Term Artificial Intelligence Coined 70 Years Ago
MIT CSAIL post recalls the conference proposal that first used the term artificial intelligence.
19. Vaibhav Srivastav Has AI Agents Email Video Demos with PRs
OpenAI engineer shares a workflow using AI agents to send video demos for review.
20. Chollet Describes Two Axes of Test-Time Scaling
Researcher notes breadth as key alongside depth for hard search problems.