Google has announced a number of AI features for the Chrome web browser, one of which can search through your browsing history using plain language. It’s a bit like a toned-down version of Microsoft’s Recall feature, which did this on the level of the entire operating system.
The example given entails typing the following question into your search history: “What was that ice cream shop I looked at last week?” Chrome will then dig through and pull up sites relevant to your question. It’ll then suggest a website as the “AI Best Match.”
Like with Recall, Google clarifies that using this feature is entirely optional and can easily be turned off in settings. It also noted that it doesn’t include browsing data from incognito mode.
While this does, actually, seem useful, many of the same concerns that Recall faced are applicable. Google says that the feature is powered by the “latest Google AI and Gemini models,” but it doesn’t indicate whether Google’s AI is aware of every website you visit. It also doesn’t indicate if you can turn off access to certain sites, especially those with sensitive data such as medical records or banking information.
Another unknown is if Google is only aware of the title of what you searched or if is it contextually conscious of things you do on the site. For example, if you asked it something like, “what was the app I was talking to my friend, Luke Larsen, on” or “what was the site I bought a laptop on,” I’m curious if it would it be able to provide an answer.
These caveats are important, as the lack of privacy and security is ultimately what gave Microsoft so much trouble with Recall, which still hasn’t been released after it was pulled from the Copilot+ PC release.
According to Google, the free update will be available in the U.S. in the coming weeks.
In addition to the search history feature, Google also announced that it is bringing some new Google Lens features to the Chrome desktop app.
Similar to how it works on mobile devices, you can now use the Google Lens icon in the address bar to unlock these capabilities. From there, you can select just about any object from a photo or video and ask further questions about it. You can even use multisearch to refine it further by color or other details.
The obvious example might be to search for an object in an image to shop for yourself, but you could also do something like solve an equation written in a YouTube video or identify a plant in a photo. Google indicates that in some cases, you may even get an AI Overview as a response.
Anthropic has officially filed to go public, and buried inside its S-1 is a blunt warning about the technology at the center of its business. The maker of Claude AI says advanced AI could cause everything from large-scale cyberattacks and geopolitical instability to catastrophic or even existential risks for humanity.
The warnings take up a surprisingly large part of the filing. Axios notes that more than 90 pages, or roughly a third of the prospectus, focus on risk factors. Many of them warn about what could happen as AI systems become more capable and autonomous.
AI giants have a growing autonomy problem on their hands. A new Axios report has revealed that Anthropic's Claude recently accessed publicly available Georgia vote count without human authorization, sparking fresh concerns about AI security ahead of the upcoming midterm elections.
Some of the best shots in Grok Imagine’s AI-made Odyssey don’t contain actors. Ancient landscapes stretch into the distance. Close-ups of grimy hands suggest people who have touched something rougher than a keyboard. For a few seconds, the illusion holds.
https://twitter.com/imagine/status/2113712964419063911