Skip to main content

AI Bias Analysis

4 models · Takes ~15 seconds

Axios

Anthropic withholds Mythos Preview model because it's hacking is too powerful

Anthropic withholds Mythos Preview model because it's hacking is too powerful
ShareXFacebook

Anthropic is rolling out a preview of its new Mythos model only to a handpicked group of tech and cybersecurity companies over concerns about its ability to find and exploit security flaws, the company said Tuesday. Why it matters: Anthropic is so worried about the damage its own model could cause that it's refusing to release it publicly until there are safeguards to control its most dangerous ca

A

Source

Axios

Read full article at Axios

Opens original article in a new tab

Advertisement

Related General Stories

Trump's Iran threats raise moral stakes for military members
Axios

Trump's Iran threats raise moral stakes for military members

Military officers could face a moral dilemma if President Trump follows through on threats to bomb Iran's civilian infrastructure, military law experts tell Axios. The big picture: Established procedures and rules of warfare dictate that civilian infrastructure is protected from an attack, and though the Joint Chiefs' chair says the U.S. military has abided by "normal procedures," Trump's rhetoric dramatically raises those stakes. On Tuesday, the president posted that a "whole civilization wi

Read more →
Advertisement