Gizmodo

Security researchers at IBM say they were able to successfully “hypnotize” prominent large language models like OpenAI’s ChatGPT into leaking confidential financial information, generating malicious code, encouraging users to pay ransoms, and even advising drivers to plow through red lights. The researchers were able…

Gizmodo

Prosecraft.io, a site that used novels to help power a data-driven project to display word count, passive voice, and other much more subjective, writing-style markers such as vividness, shut down today after authors protested the project.

Tech Insider
Mark Zuckerberg speaks on stage against a blue background.
Mark Zuckerberg is focusing more of Meta's efforts on commercializing AI.
Tech Insider
Photo illustration showing ChatGPT and OpenAI logos.
Just one line of code can now prevent OpenAI from accessing a website's data to train ChatGPT.
Tech Insider
The National Ignition Facility’s preamplifier module increases the laser energy as it travels to the Target Chamber
US government scientists reportedly produced an energy gain during a fusion reaction for the second time ever, in a possible breakthrough.
Gizmodo

Google is just plain tired of all your grammar flubs in Search. The company announced Monday it had added a simple grammar check feature into the world’s largest search engine, though you won’t be using it to shore up any writing you want to show someone else.

Read more...

Tech Insider
Godfathers of AI
Three of the "godfathers of AI" helped spark the revolution that's making its way through the tech industry — and all of society. They are, from left, Yann LeCun, Geoffrey Hinton, and Yoshua Bengio.
Gizmodo

By now, just about everyone has seen the digital art and prose of generative AI models like Midjounrey and ChatGPT.

Gizmodo

In the latest chapter of Elon Musk’s saga to take over the world with corporations named after the letter “X,” the billionaire tech tycoon just snagged a coveted URL: AI.com.

Read more...