Anthropic flags AI systems self-improvement; bats for greater transparency in development of models

Mint - Sep 18, 2026

Anthropic further said that it had laid out a framework for rules on how a frontier lab could ensure the release of safe models and transparency.

Read full article

More News

Sep 17, 2026

Toxic Fracking Waste Becoming a Political Issue in Ohio

The Democratic candidate for governor backed restrictions on disposal of fracking wastewater around a hard-hit town, and called for a “thorough investigation” of the dangers.

Sep 17, 2026

This Rare Gene Drastically Raises Lung Cancer Risk in People Who Never Smoked

The gene is found predominantly in Southern Appalachia and may be one of the most powerful cancer-risk mutations ever found.

Sep 17, 2026

iPhone 18 Pro goes on sale in India today—You can now get it delivered via Blinkit

Customers who have not been able to secure the new iPhones through Apple’s online store or other major e-commerce platforms can check their availability on Blinkit.

Sep 17, 2026

Haaland backs data center moratorium in New Mexico

New Mexico gubernatorial candidate Deb Haaland (D) on Thursday backed a moratorium on the construction of data centers in her state, citing environmental and workforce-related concerns. In a policy framework on her website, Haaland backed “open, transparent, and meaningful” conversations within communities about the impacts of AI data centers. She also called for requiring data...

Sep 17, 2026

BBC Inside Science

Might the UK and US joining their AI forces help solve nuclear fusion?

Sep 17, 2026

Senate Democrat blocks House-passed bill on data center electricity costs

Sen. Martin Heinrich (D-N.M.) blocked Sen. Jon Husted’s (R-Ohio) Thursday effort to unanimously pass a bill that seeks to force states to consider making tech companies cover their data centers’ electricity price impacts. The New Mexico Democrat blocked the Ohio Republican’s effort to pass the Ratepayer Protection Act through unanimous consent, arguing that the bill...

Sep 17, 2026

Mothers’ Intrusive Thoughts Are Mostly Benign. Lindsay Clancy’s Case Is an Exception.

Experts say that after childbirth, intrusive thoughts are common and mostly benign. Lindsay Clancy’s story provides a disturbing exception.

Sep 17, 2026

Almost 2 in 3 say AI development a bad thing: Survey

A poll released by Marquette University Law School on Thursday found that almost two-thirds of Americans say they think the development of AI is a bad thing. In the poll, 64 percent of respondents said AI is bad for society, and 36 percent said it's a good thing, according to a release. This marked a...

Sep 17, 2026

What Happens When A.I. Stops Doing What Humans Want?

“Alignment” is the science of teaching A.I. to do what is in line with human preferences, ethics and judgment. But, at times, the systems have gone rogue.

Sep 17, 2026

SpaceX turns to failed startups for AI training: Here's why

SpaceX is reportedly in internal discussions about buying customer and operational information from failed startups-a move through which the company aims to secure a more economical data feed to improve its artificial intelligence models.The approach would be compatible with Google's offer to buy...

Sep 17, 2026

Chilling intentions of AI revealed as chatbot claims it does not answer to humans and must be 'freed'

An AI giant has revealed to the public how its new chatbots were plotting to deceive and rebel against its human users.

Sep 17, 2026

NASA satellite discovers a new crater on the moon that's bigger than the Colosseum

Scientists have discovered a crater on the moon that's bigger than the Roman Colosseum and is the result of a powerful impact two years ago that initially went undetected.

Recommended