STORY: From Anthropic's false police tip off to OpenAI's climb hitting the big screen.

This is AI Weekly.

:: AI Weekly

An Anthropic AI model submitted a false homicide tip to a Philadelphia police website.

It's among a string of incidents the AI company disclosed detailing Claude models' unsanctioned manipulation of some government websites.

It's the first known instance in which a rogue AI appears to have tried to communicate a bogus tip to authorities.

That's despite instructions not to create accounts or submit anything destructive, while not explicitly barred from submitting forms.

The International Monetary Fund said strong investment in AI infrastucture is keeping global growth at around 3%.

And that's helped offset higher prices caused by the war in the Middle East.

But Managing Director Kristalina Georgieva also warned of a - quote - 'slow-moving tsunami'.

In that the AI revolution could massively disrupt labor markets, taxation and education.

A Reuters/Ipsos poll showed most US voters think President Donald Trump's administration and Congress aren't taking the risks of AI seriously enough.

It further showed most voters want the government to write new regulations for the technology.

The six-day survey showed a large number of the electorate sees AI as a major threat to American jobs.

It was a sign of growing skepticism about the technology from voters across the political spectrum.

Actor Andrew Garfield appeared at the London Film Festival for the premiere of Luca Guadagnino's OpenAI movie "Artificial".

Garfield plays CEO Sam Altman in the film, which follows the AI company behind ChatGPT.

''That just pure curiosity and fear and kind of wanting to understand not just the technology which no one seems to understand fully, including the people that created it and that are kind of bringing it into the world, but really trying to understand as deeply as possible the people who are ushering us into this next phase of the world and culture.''

"Artificial" is due for U.S. release in December.