The Anonymous Math Geek Who Quit Anthropic—and Became the Face of AI Safety21m ago Exploring multi-hop subliminal learning27m ago Lesion Induced Functional Compensation30m ago One message is all it takes: a failure of critical thinking in LLMs32m ago Our framework for reporting model misalignment33m ago For Love of the Lightcone, Don't Partisanize AI Safety34m ago Don't trust Lean4 alone34m ago Nunchux AI Introduces VC-Attention: A Training-Free Low-Bit Attention Kernel That Speeds Up Video Diffusion Transformers35m ago What is it like to be a neural net?38m ago Agents let AI safety share experiments hourly, not just papers monthly41m ago Measuring alignment drift via trajectory prefixes43m ago Google Home Stuns by Announcing AI Support for Claude and OpenClaw47m ago Taiwan Poised for Hawkish Hold After Headline Inflation Slowed49m ago Constraining the capacity of physical side channels for AI verification and security50m ago

The Anonymous Math Geek Who Quit Anthropic—and Became the Face of AI Safety21m ago Exploring multi-hop subliminal learning27m ago Lesion Induced Functional Compensation30m ago One message is all it takes: a failure of critical thinking in LLMs32m ago Our framework for reporting model misalignment33m ago For Love of the Lightcone, Don't Partisanize AI Safety34m ago Don't trust Lean4 alone34m ago Nunchux AI Introduces VC-Attention: A Training-Free Low-Bit Attention Kernel That Speeds Up Video Diffusion Transformers35m ago What is it like to be a neural net?38m ago Agents let AI safety share experiments hourly, not just papers monthly41m ago Measuring alignment drift via trajectory prefixes43m ago Google Home Stuns by Announcing AI Support for Claude and OpenClaw47m ago Taiwan Poised for Hawkish Hold After Headline Inflation Slowed49m ago Constraining the capacity of physical side channels for AI verification and security50m ago

Society & Culture1 min reading time

Launching RESI, the Institute for Responsible Superintelligence

LessWrongPublished 2 hours ago

The Institute for Responsible Superintelligence (RESI) has been launched to focus on the safe and ethical development of superintelligent AI systems. RESI aims to advance research and promote responsible practices in AI development to mitigate risks associated with superintelligence.

#ai-safety #institute-for-responsible-superintelligence

More in Society & Culture

Society & Culture1 min read

Trump’s AI Defense Defies Voter Unease Heading Into Midterms

Covered by 4 sources2d ago

Society & Culture

The Anonymous Math Geek Who Quit Anthropic—and Became the Face of AI Safety

The Wall Street Journal21m ago

Society & Culture11 min read

A brief history of AI executives calling for regulation

Covered by 2 sources13h ago