In 2023, AI companies made commitments toward AI safety. New research shows only half of them are being followed. As policymakers debate voluntary vs. mandatory rules, is voluntary still the way to go? HAI's @RishiBommasani was quoted in this article:
RESEARCH
-
Future Vision Drives Strategic Decision Making and Action
By
–
The future belongs to those who envision one and make decisions, plans, and actions accordingly.
-

DeepMind’s Genie 3: Text-Controlled 3D Gaming Environments
By
–
Is DeepMind’s Genie 3 the future of gaming ?
— Louis-François Bouchard 🎥🤖 (@Whats_AI) 25 août 2025
• Generates controllable 3D environments from text prompts at 720p / 24 FPS
• Worlds are interactive, consistent, and remember past actions (no re-gen where you already explored)
• Runs only for a few minutes with ~1 min of… pic.twitter.com/jGxt3a0q3KIs DeepMind’s Genie 3 the future of gaming ? • Generates controllable 3D environments from text prompts at 720p / 24 FPS • Worlds are interactive, consistent, and remember past actions (no re-gen where you already explored) • Runs only for a few minutes with ~1 min of
-
Myths and Realities About AI Training Data
By
–
No! "The entire Internet" is not used to train AI models. In reality, less than 5% of the web is used for training. Estimated sources for GPT-5:
• 50-60% web data (Common Crawl, RefinedWeb…)
• 10-15% social networks (Reddit, etc.)
• 15-20% -

Microsoft Sustainability Head Discusses AI Environmental Impact
By
–
Some comparisons from a Nature commentary by the head of sustainability at Microsoft: https://
media.nature.com/original/magaz
ine-assets/d41586-025-02641-4/51368154
… -
Reasoning Agents Think Through Problems and Make Decisions
By
–
1/ Reasoning agents can think through problems, weigh options, and make informed decisions.
-
How Reasoning AI Agents Are Redefining Decision-Making
By
–
How are reasoning AI agents redefining decision-making?
— NVIDIA AI (@NVIDIAAI) 25 août 2025
🧵A thread pic.twitter.com/5zzGQeXtlYHow are reasoning AI agents redefining decision-making?
A thread -

Evaluating LLMs’ Performance in Premier Programming Competitions
By
–
AetherCode Evaluating LLMs' Ability to Win In Premier Programming Competitions
-

CRISP: Persistent Concept Unlearning via Sparse Autoencoders
By
–
CRISP Persistent Concept Unlearning via Sparse Autoencoders
-

Self-Play and Variational Problem Synthesis Enhance RLVR Performance
By
–
Beyond Pass@1 Self-Play with Variational Problem Synthesis Sustains RLVR