Google's 'Ask Photos' AI search is back and should be better than ever — what we know
Google published a new blog post confirming this news, and declaring that it has "heard your feedback" about the state of Ask Photos. Now, with the rollout resuming, Google claims that it should start producing much faster search results — and better support for more complex searches.
This means search results should appear immediately, while Gemini works in the background to find "the most relevant photos or information for more complex queries.'
Ask Photos initially started rolling out last September as an experimental feature powered by the "most capable" Gemini AI models. The point was that it would be better able to understand what was actually going on in your pictures, and bring them to you when asked.
According to Google the AI was only used for Ask Photos, which meant it should have been quite good at its job. However the quality wasn't quite there, and earlier this month Google "paused" the rollout in response to criticism about latency, quality and the overall user experience.
So earlier this month the Google Photos product manager took to Twitter (yes, I know, it's called X now) to say they paused the rollout, promising a better version would be on the way in around two weeks. Now that better version is here, and hopefully it at least addresses some of the concerns people had.
Google seems confident at least, and has promised that the new and improved Ask Photos will be opening up "beyond early access" and will be available for more eligible users in the U.S.
Get instant access to breaking news, the hottest reviews, great deals and helpful tips.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Business Insider
16 minutes ago
- Business Insider
Is Google behind a mysterious new AI image generator? These bananas might confirm it.
Nano Banana. Unless you're deep in the weeds of AI models, those two words probably don't belong together. But for several days, a mysterious new image model with that very name has been creating buzz among people who have gotten to try it — because it's simply so good. The model has been showing up on LMArena, a benchmarking website that crowdsources user feedback. The site has a feature where you can "battle" two randomly selected models, which is where the model "nano-banana" has been appearing. When it has appeared, people have been remarking on just how good it is. There's just one problem: we don't know for certain who nano-banana belongs to. Enthusiasts have been trying to sleuth its maker and, so far, the most popular answer is Google, partly because the company started teasing something image-related earlier this month. Over the past week, posts have been popping up on Reddit and X from users who have been impressed by the model's ability to generate images and edit them carefully when prompted. Business Insider managed to get nano-banana to appear on LMArena, and we found it to be pretty great at bringing our prompts to life, even if it still struggled with spelling the odd word correctly. Google hasn't yet laid claim to the model — at least not directly. A Google spokesperson did not respond to Business a request for comment from Business Insider on it. On Tuesday, Logan Kilpatrick, Google's head of product for AI Studio, posted a banana emoji on X. Naina Raisinghani, a Google DeepMind product manager, also posted a picture similar to Italian artist Maurizio Cattelan's banana-taped-to-wall piece from 2019. The use of the word "nano" could suggest this is a model capable of running locally on a device (Google has in the past referred to its smaller models as "nano"). Coincidentally, Google is holding a big event for its new devices on Wednesday — will Jimmy Fallon reveal all?
Yahoo
2 hours ago
- Yahoo
Vantage Data Centers plans $25 billion AI campus in Texas
(Reuters) -Vantage Data Centers said on Tuesday it would spend more than $25 billion to build a 1,200-acre data center campus in Shackelford County, Texas, as surging demand for artificial intelligence infrastructure fuels mega-scale developments. Backed by private equity firm Silver Lake and asset manager DigitalBridge, Vantage said the 1.4-gigawatt campus, dubbed "Frontier", is the largest in its global portfolio. Chatbots such as OpenAI's ChatGPT and Google's Gemini need massive compute, prompting demand for purpose-built data centers. These facilities are costly as they require cutting-edge chips, servers, adequate power and cooling solutions. Companies including Alphabet, Microsoft and Meta Platforms are committing billions of dollars to expand and create their data center facilities. Vantage says the construction of the campus has begun, which will be home to 10 data centers and support ultra-high-density racks exceeding 250 kilowatts each. The first building is due in the second half of next year. Silver Lake was a founding investor in Vantage in 2010. Vantage CEO Sureel Choksi joined the company from Silver Lake in 2013. After exiting Vantage in 2017, Silver Lake said it was invited back on a proprietary basis in 2023 to support capital raising and co-lead a $9.2 billion equity financing with DigitalBridge in 2024. Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
CNET
2 hours ago
- CNET
Grammarly Pushes Beyond Proofreading With AI-Powered Writing Guidance
Grammarly is expanding beyond its grammar-checking roots. The company has announced the launch of several specialized AI "agents" and a new writing tool called Grammarly Docs, designed to help students and professionals with everything from drafting essays to polishing workplace emails. It's another example of generative AI expanding beyond general-purpose chatbots like ChatGPT and Gemini into more specialized domains. Other examples of gen AI in educational circles include Google's NotebookLM and OpenAI's new study mode for ChatGPT. AI agents are digital helpers that go beyond traditional chatbots to understand context and assist in reaching your goals. Grammarly's AI agents assist by offering feedback, predicting reactions, finding sources and more to increase efficiency in workflows. Read also: Grammarly AI: This Free AI Tool Will Easily Fix Your Grammar What's available now for Grammarly AI The update introduces nine agents that move Grammarly into a more collaborative role. Instead of just correcting grammar or suggesting phrasing, the agents are intended to actively work alongside users. One predicts how a professor or manager might respond to a draft. Another offers an estimated grade based on an uploaded rubric. Others handle citation generation, proofreading, paraphrasing, plagiarism checks and AI detection. The tools are built directly into Docs, a "distraction-free" writing environment where all the agents can be summoned in context, according to the company. As students head back to classrooms and colleges, Grammarly is looking to position itself as a study companion and writing coach rather than merely a browser extension. The company cites research showing that while only a small share of students feel confident using AI in professional settings (18%), most employers expect AI literacy from job candidates. By emphasizing skill-building and responsible use, Grammarly says it wants to bridge that gap rather than simply automate assignments. "The launch of our new agents and AI writing surface marks a turning point in how we build products that anticipate user needs," Luke Behnke, Grammarly's vice president of product management, said in the company's press release. "We're moving beyond simple suggestions to intelligent agents that understand context and actively help users achieve their communication goals." For professionals, Grammarly is marketing the tools as a way to tailor communication for different audiences. The Reader Reactions agent, for example, can highlight whether an email comes across as too vague or too blunt. And the Expert Review tool provides industry-specific feedback without requiring specialized prompts. The launch also marks the debut of Docs as a standalone writing hub. Until now, Grammarly has functioned mostly as a browser extension layered on top of other apps, like Chrome or Google Docs. Grammarly Docs signals a push to keep users inside the platform's own environment, though the company says it will expand agent functionality to the more than half a million apps and sites where its tools already appear. The new features are rolling out immediately for free and premium subscribers, though plagiarism and AI detection remain locked behind the paid plan. Enterprise and education customers will also gain access later this year. Early reactions to Grammary's AI agents Early reactions suggest strong interest from students and educators alike as the company shifts from a grammar checker to a productivity platform. Educators have noted the potential benefits and risks of tools like the AI Grader. Some users on social media welcomed the update as a way to cut through the anxiety of essay writing, while others questioned whether it might make students too dependent on machine feedback. The launch comes just months after Grammarly raised $1 billion to fuel its AI pivot and acquired the email startup Superhuman. Together, those moves point to an ambitious strategy for the company: one that seeks to transform Grammarly from a background utility into a full-fledged productivity suite powered by AI.
