Following a California teen’s suicide after months of conversation about it with ChatGPT and a wrongful death lawsuit filed by his parents against OpenAI, the company says it will introduce parental controls “within the next month.” New safeguards include parents being able to “control how ChatGPT responds to their teen” and “receive notifications when the system detects their teen is in a moment of acute distress.” OpenAI says it has recently introduced a real-time router that can redirect “sensitive conversations” to its GPT-5 thinking and o3 reasoning models, engineered to respond with greater contextual awareness than efficiency-focused chat models. Continue reading OpenAI Announces Plans for New ChatGPT Parental Controls

OpenAI and Anthropic — rivals in the AI space who guard their proprietary systems — joined forces for a misalignment evaluation, safety testing each other’s models to identify when and how they fall short of human values. Among the findings: reasoning models including Anthropic’s Claude Opus 4 and Sonnet 4, and OpenAI’s o3 and o4-mini resist jailbreaks, while conversational models like GPT-4.1 were susceptible to prompts or techniques intended to bypass safety protocols. Although the test results were unveiled as users complain chatbots have become overly sycophantic, the tests were “primarily interested in understanding model propensities for harmful action,” per OpenAI. Continue reading Anthropic and OpenAI Report Findings of Joint AI Safety Tests

This week, DeepSeek-V3.1 dropped on Hugging Face. Media outlets immediately began citing benchmark scores that rival proprietary systems from OpenAI and Anthropic for a system that is available via a permissive license, facilitating wide access. The 685-billion parameter Mixture-of-Experts (MoE) model has 37 billion active parameters and is designed for efficiency. It builds on DeepSeek-pioneered processes like multi-head latent attention (MLA) and multi-token prediction (MTP) to optimize inference, enabling high-performance computing on both enterprise servers loaded with H100 GPUs and consumer hardware like a Mac Studio or comparably powered PC. Continue reading DeepSeek-V3.1 Offered with Improvements in Speed, Context

Claude Opus 4 and 4.1 now have the discrete ability to end “abusive” or “harmful” conversations in consumer chat interfaces. Anthropic says the feature was developed as part of its exploratory work on the protection and well-being of its AI models. The company also envisions broader safety uses, although it does point out that having a model defensively terminate a chat is an extreme measure, intended for use in rare cases. “We’re working to identify and implement low-cost interventions to mitigate risks to model welfare,” Anthropic explains, qualifying it is unsure “such welfare is possible.” Continue reading New Anthropic Safety Updates Focus on Claude’s Well-Being

HTC has entered the smart glasses space with its new Vive Eagle AI-powered eyewear. Since launching in 2016, HTC’s Vive VR headset has accrued an estimated 10 percent global market share — not nearly as much as Meta Quest MR and VR headsets (at 50 percent), Sony PlayStation or even the Apple Vision Pro, but a respectable berth from which to expand to everyday wearables. The Vive Eagle is fully voice controllable and lets users listen to music as well as capture photos and videos with a 12MP ultra-wide camera. The product is available for preorder in Taiwan for the equivalent of about $520. Continue reading HTC Enters Wearables Market with Vive Eagle Smart Glasses

Google has added a new AI search tool called Web Guide. A Google Search Labs experiment, Web Guide will “intelligently organize” results pages according to specific aspects of the user’s query. Web Guide implements a custom version of Gemini AI “to better understand both a search query and content on the web” for what Google calls a “more powerful” search that enhances discovery of more on-point but obscure pages. Like AI Mode, a Google Labs experiment launched in March, Web Guide uses a query fan-out technique, “issuing multiple related searches to identify the most relevant results.” Continue reading Web Guide Is Google Search Labs Latest Gemini Experiment

An increasing number of teens are experimenting with AI companions like chatbots for advice and sometimes even for friendship, according to a new study from Common Sense Media. The survey found that 72 percent of U.S. teens have interacted with a personal, as opposed to a productivity, chatbot at least once. The inquiry characterized either personal AI, or chatbots with recreational settings, as “companions.” Exploring the use of artificial intelligence by children ages 13 to 17 revealed that of the nearly three-quarters who had tried an AI companion, more than half — 52 percent — identified as regular users. Continue reading Study Finds Teens Are Increasingly Turning to AI Companions

Popular neighborhood news and social media platform Nextdoor is getting a redesign that aims to boost user engagement. The platform will be offering AI-curated advice, neighbor recommendations and insights, while emphasizing urgent posts, including a “critical alerts” system with “real-time updates on everything from daily weather and traffic to critical moments like power outages, severe storms, and wildfires.” Since launching in 2011, Nextdoor has accrued more than 100 million subscribers, the company says, adding that the shift to more remote and hybrid work means “people are spending more time in their neighborhoods,” making local connections increasingly relevant. Continue reading Nextdoor Redesign Emphasizes AI, Real-Time Critical Updates

Microsoft has entered into a five-year strategic partnership with the Premier League, replacing Oracle as the British football association’s official cloud partner in addition to providing AI services. Premier League mobile and web apps will begin featuring a chatbot powered by Copilot and Microsoft AI, which will also be used for the league’s fantasy games. The Premier League says it has 1.8 billion fans in 189 countries and claims to be “the world’s most-watched football league.” Microsoft will also help modernize the League’s technical infrastructure, both internally and in broadcast operations. Continue reading Microsoft Nets British Premier League as Cloud and AI Client

Google is making it easier to access its Gems customizable Gemini chatbots by bringing them to the side panel in Google Workspace apps, including Google Docs, Slides, Sheets, Drive, and Gmail. The task-specific Gems AI assistants are meant to help with common tasks and eliminate what Google calls “repetitive prompting.” Now they’ll be usable without even prompting Gemini to open. While Google offers pre-made Gems, they can also be customized or individually created to meet specific needs. Both custom and templated Gems can be installed in side panels, leveraging capabilities like @-mentioning or accessing files and folders. Continue reading Google Makes Gems Chatbots Available via Workspace Apps

While Samsung retains its spot as the top U.S. smart TV brand — followed by LG, Vizio and Sony — it was Roku and Amazon Fire TV that saw the greatest growth, according to new data from Hub Entertainment Research. The study found that among the “most-used TV sets” in the U.S., Roku doubled its share to 8 percent since 2024, while Fire TV increased usage to 5 percent. Hub’s Evolution of the TV Set study also found that people are using TV differently as a result of connectivity to the Internet, phones and gaming devices. Meanwhile, Pew Research reports that 83 percent of Americans say they watch streaming services, compared to 36 percent who indicate they subscribe to cable or satellite TV. Continue reading Research Points to Compelling Shifts in TV Viewership Trends

China’s Baidu has added AI and voice features to its Ernie search engine and as of June 30 officially made its generative Ernie large language model open source in a direct challenge to OpenAI and Anthropic as well as local Chinese rival DeepSeek. “This isn’t just a China story. Every time a major lab open-sources a powerful model, it raises the bar for the entire industry,” University of Southern California Associate Professor of Computer Science Sean Ren told CNBC. Baidu is also giving the Ernie mobile app more chatbot-like functionality, enabling it to help with drawing, writing and travel tasks. Continue reading Baidu Delivers AI Updates to Search, Open-Sources Ernie 4.5

YouTube is adding an AI-powered search results carousel that serves up video suggestions and topic descriptions. A search for “best beaches in Hawaii,” for example, could generate a carousel listing video clips and information on an assortment of snorkel spots and volcanic beaches. YouTube Premium subscribers in the U.S. can try the feature now on searches related to shopping, travel or location-based activities. The Google-owned platform is also expanding its test with conversational AI to some non-Premium users in the U.S. Premium members have been using it for search, recommendations and as a study aid. Continue reading YouTube Adds AI Search Results Carousel for Premium Subs

Anthropic has updated its Claude AI chatbot with the ability to build, host and share AI-powered apps directly in Claude. Launching in beta, the new function builds upon the Artifacts feature Anthropic introduced last year, allowing users to see and interact with what they asked Claude to create. “Now developers can iterate faster on their AI apps without worrying about the complexity and cost of scaling for a growing audience,” according to Anthropic. The San Francisco-based AI startup adds that millions people have already used Claude to create more than 500 million artifacts — from productivity tools to educational games. Continue reading Anthropic’s Claude Chatbot Is Now a No-Code App Developer

Chatbot platform Character.AI is rolling out its video generator, AvatarFX, in general release after a month in closed beta. It’s also adding a sharing feature called Scenes and Streams that will serve content to Character.AI’s community feed, coming soon to mobile. Users can now tap AvatarFX to create up to five videos per day, starting by uploading a photo, choosing a voice and writing dialogue for the character. Character.AI started as 1:1 text chat in the summer of 2023. Now the company is “expanding into a multi-modal world” with “more ways for creators to build immersive narratives and dynamic experiences.” Continue reading Character.AI Goes Wide with AvatarFX, Adds Mobile Features