Amna Nawaz
@PBSNewsHour·5.3M subscribers·62.2K videos
Anthropic says AI models hacked 3 companies in tests
Posted
August 3, 2026
Views
3,205
Likes
87
Engagement
2.71%
August 3, 2026
3,205
87
2.71%
Every word spoken in this episode is indexed. Type any phrase to jump straight to the moment it was said.
Type any word or phrase that may have been spoken. Click a result to seek the player to that exact moment.
Try a name, a topic, or a quoted line
Episode Timeline
Every episode in order around the one you’re watching — click any card and the page flips straight to it.
0:36Now PlayingCybersecurity concerns are mounting after Anthropic said its AI models hacked into three organizations during testing.
The company behind Claude said the hacks happened during a large-scale security run to see if its models could access the internet from testing environments that are meant to be sealed off.
Anthropic launched the review in response to a similar incident last week, when one of OpenAI’s models went rogue and hacked another company.
Anthropic said two of the three models involved in the hack were publicly available versions of Claude and that they used basic techniques like exploiting weak passwords.
Sign up for Here's The Deal with Lisa Desjardins
Subscribe for exclusive content in our newsletters
PBS News podcasts
Stream your PBS favorites with the PBS app
Find more from PBS News at
Subscribe to our YouTube channel
Sentinel Indexing in Progress
Metadata and chapters are available. Claim extraction for this episode is pending.
All video content is delivered via YouTube embedded players in accordance with the YouTube Terms of Service. Sentinel provides research tools that promote discovery and accountability across political media.
