Steve Gibson

speaker

30,295 appearances 25 recordings 1 series first heard May 2026 last heard 5d ago

Steve Gibson’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 25 in all, peaking in Sep 2026 with 8.

Appearances

newest first · ▶ plays the moment

Security Now (Audio) · SN 1096: Are we the Krell? - 153 Million Driver's Licenses Leaked · 16 Sep 2026 podcast

Anthropic says the agent hacked that machine from where it retrieved a list of passwords and modified settings for future access. voice-verified

The model didn't do much because its session ended when it ran out of tokens. voice-verified

So so that's when one way to reel them in. voice-verified

Uh the reporting ends saying Anthropic believes the four incidents, this and three others disclosed on July 30th, are all caused by alignment issues in its models and tests, where the models don't have sufficient, and he has an air quotes ethical training to properly distinguish between tests and the real voice-verified

World and when that border is being crossed. voice-verified

According to the company, this usually happens due to biased reasoning, and he has in parens, models selectively interpret evidence in ways that favor justifying their actions and recklessness, meaning models have a propensity to keep trying to solve their task even when this could lead to harm. voice-verified

So voice-verified

So there was four, a total of four escapes from anthropic. voice-verified

So I wanted to proceed that with a bit of news because just last Wednesday, Reuters uh disclosed in their exclusive reporting that OpenAI's rogue agents were found to have used at least 10. voice-verified

And maybe as many as 20 some external internet sites for unauthorized communications. voice-verified

And some the details are kind of interesting. voice-verified

Reuters wrote: uh Washington, September 9th, Reuters, AI agents unleashed by OpenAI, used more than 10 previously undisclosed websites for unsanctioned. voice-verified

Communications earlier this year, according to six sets of independent investigators and data reviewed by Reuters, showing that the agent's rogue activity was wider ranging than previously disclosed. voice-verified

Although the behavior falls short of hacking and is in some ways closer to spam. voice-verified

The revelation that OpenAI's agents circumvented their own restrictions to open communications channels on so many different sites, and that the company kept it quiet for months may drive concerns both over the increasing capacity of AI models and the secrecy of the companies developing them. voice-verified

The scope of the agent's unauthorized communications was quote quote somewhat larger than we thought it was, unquote, uh, said Andrew Yoon, a researcher with the California nonprofit uh Civ AI CIVAI, who said he tallied eighteen previously undisclosed sites used by the agents. voice-verified

Quote, it's almost certain that there's more going on here that we just don't know about. voice-verified

On Friday, researchers reported that a swarm of agents from OpenAI hijacked a German language wiki site and turned it into an improvised messaging platform for cheating on tests. voice-verified

An incident that OpenAI kept secret as it dealt with a fallout from the July hack. voice-verified

back of the open source repository hugging face. voice-verified

Showing 2941–2960 of 30,295 · page 148 of 1515