Umjetna inteligencija (AI)

[tomek@vz]

When it comes to AI, Google has left its "do no harm" stance firmly in the past. The Pentagon has just announced GenAI.mil, a generative AI platform that will offer Google Cloud's Gemini as its first available AI tool. Secretary of Defense/War Pete Hegseth said that the platform will make "our fighting force more lethal than ever before."

"The future of American warfare is here, and it's spelled AI," Hegseth said. "As technologies advance, so do our adversaries. But here at the War Department, we are not sitting idly by."

> Pentagon says its new military AI platform with Google's Gemini will make US forces "more lethal"

Companies that are excited about generative AI speeding up productivity – usually at the cost of human employees – often don't care about the public's opinion. But McDonald's just found out what happens when you ignore the general consensus. Its new AI-made Christmas ad has been pulled after being universally panned.

McDonald's Netherlands unwisely decided that an AI-generated ad would be a good idea to promote its Christmas campaign. It's pretty easy to spot the tell-tale signs that something was created by an AI, especially when it includes "real" people, as this 45-second ad does.

> Watch the AI-generated Christmas ad McDonald's pulled after massive backlash

[Promo]

https://www.youtube.com/shorts/T8TCN0aYs2s

[tomek@vz]

Stanford researchers spent much of the past year building an AI bot called Artemis that scans networks for software vulnerabilities, and when they pitted it against ten professional penetration testers on the university's own engineering network, the bot outperformed nine of them. The experiment offers a window into how rapidly AI hacking tools have improved after years of underwhelming performance.

"We thought it would probably be below average," said Justin Lin, a Stanford cybersecurity researcher. Artemis found bugs at a fraction of human cost -- just under $60 per hour compared to the $2,000 to $2,500 per day that professional pen testers typically charge. But its performance wasn't flawless. About 18% of its bug reports were false positives, and it completely missed an obvious vulnerability on a webpage that most human testers caught. In one case, Artemis found a bug on an outdated page that didn't render in standard browsers; it used a command-line tool called Curl instead of Chrome or Firefox.

Dan Boneh, a Stanford computer science professor who advised the researchers, noted that vast amounts of software shipped without being vetted by LLMs could now be at risk. "We're in this moment of time where many actors can increase their productivity to find bugs at an extreme scale," said Jacob Klein, head of threat intelligence at Anthropic.

>https://it.slashdot.org/story/25/12/...beating-humans

[Neo-ST]

Huge if big.
Imaš link na tog Artemisa da isprobamo?