Google Cloud is rolling out its seventh-generation Tensor Processing Unit (TPU), Ironwood, and new Arm-based computing options that aim to meet exploding demand for AI model deployment in what the Alphabet company describes as a business shift from training models to serving end users. “Constantly shifting model architectures, the rise of agentic workflows, plus near-exponential growth in demand for compute, define this new age of inference,” explains Google Cloud. The company said that Anthropic — known for its Claude family of large language models — “plans to access up to 1 million” of the new TPUs. The deal is reportedly “worth billions.” Continue reading Google Unveils New AI Chips, Announces Deal with Anthropic

San Francisco-based AI startup Anthropic has released a smaller, cheaper Claude model called Haiku 4.5 that can reportedly outperform larger models that are only a few months old. In terms of coding skills, Haiku 4.5 performs comparably to Claude Sonnet 4 and OpenAI’s GPT-5, per the SWE-bench Verified leaderboard. “What was recently at the frontier is now cheaper and faster,” Anthropic suggests, noting that just “five months ago, Claude Sonnet 4 was a state-of-the-art model” and “today, Claude Haiku 4.5 gives you similar levels of coding performance but at one-third the cost and more than twice the speed.” Continue reading Anthropic’s New Claude Haiku 4.5 is Available to Free Users

The Wikimedia Foundation, which runs Wikipedia, has noticed an increase in bot activity on the encyclopedic website, and a decline of approximately 8 percent in human page views. The non-profit says the trend is due to a combination of the impact of social media and generative AI as informational sources for potential visitors. AI now provides summarized answers to user questions, while younger people are increasingly finding information through social video. The Foundation’s organizers say the audience loss could have a negative effect on Wikipedia donors as well as the human volunteers who keep the site current. Continue reading AI Search and Social Video Have Impact on Wikipedia Traffic

Anthropic has introduced a new general feature called Skills that lets Claude AI access specialized expertise on demand to more efficiently deploy agents in enterprise workflows. “Claude will only access a skill when it’s relevant to the task at hand,” like when working with Excel or following an organization’s brand guidelines, Anthropic explains. Skills are folders that include instructions, scripts and resources that Claude can load as needed. And users can also build their own for use across Claude apps, Claude Code, the Claude Agent SDK and the Anthropic API. Anthropic joins tech giants such as OpenAI, Google and Microsoft in efforts to make AI agents more practically useful. Continue reading Anthropic Adds Prebuilt and Custom Agent ‘Skills’ to Claude

Anthropic and IBM have formed a strategic alliance that will see Anthropic’s Claude models integrated into select IBM software products, the first of which is a new AI-first integrated development environment (IDE) called Project Bob, now in private preview, which aims to help enterprise clients do things like modernize software. At its TechXchange 2025 conference in Orlando, IBM also debuted a new capability for AgentOps. The AgentOps practices unveiled at IBM Think in May was showcased here as a real-time governance and observability integration to help ensure autonomous agents don’t run amok. Continue reading IBM Previews Its Project Bob IDE, Intros Real-Time AgentOps

Anthropic has released its latest frontier AI model, Claude Sonnet 4.5, built for high-performance coding and building complex agents. Anthropic claims it is “the best coding model in the world” and “the best model at using computers,” with substantial gains in reasoning and math. Claude Sonnet 4.5 is distinctive in that it can build “production-ready” applications and is not limited to prototypes, Anthropic says, describing it as a major gain over previous models. The new model also excels in research as well as enterprise basics like cybersecurity and finance, per Anthropic. Continue reading Anthropic Claude Sonnet 4.5 Good at Coding and Collegiality

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

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

OpenAI is releasing two lower-cost, open-weight reasoning models in an effort to be more competitive with Meta, Mistral and DeepSeek and they will be the first OpenAI models available from Amazon. The new offerings — gpt-oss-120b and gpt-oss-20b — will be among the model choices on AWS’s Bedrock and SageMaker AI services. Both models are said to be well-suited for agentic use. The gpt-oss-120b model performs comparably to OpenAI o4-mini on core reasoning and can run on a single 80GB GPU. The gpt-oss-20b model is compared to OpenAI o3‑mini and can run on edge devices with just 16GB of memory. Continue reading Open-Weight Models Are a First from OpenAI in AWS Catalog

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

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

An internal OpenAI document revealed as part of a discovery demand in the U.S. Justice Department’s antitrust case against Google reveals OpenAI’s plan to turn ChatGPT into the public’s primary Internet interface. Although heavily redacted, the document effectively reveals OpenAI’s goal is to have ChatGPT replace Google Chrome and Search with an “AI super assistant that deeply understands you and is your interface to the Internet.” Aside from bolstering Google’s position that it has competition and is not an impregnable monopoly, the information contained therein provides an in-depth look at OpenAI’s roadmap. Continue reading OpenAI Seeks to Make ChatGPT New Gateway to the Internet

Anthropic’s new mobile conversation voice mode for its large language model Claude lets it search Google Docs, Drive, Calendar and more on smartphones. Just a week after debuting two new LLMs — Claude Opus 4 and Sonnet 4 — Anthropic announced the mobile updates for its Claude AI chatbot for iOS and Android and said it is extending web search for all users on free Claude plans. While Claude’s conversational voice interface is currently available only in English and only via mobile, an API for desktop use and browser-based support are part of future plans. Amazon and Google both have investment stakes in San Francisco-based Anthropic. Continue reading Anthropic Touts Mobile Voice Mode, Free Search for Claude

Apple has teamed with Anthropic on a “vibe coding” AI platform that will write, edit and test software for developers. The system is essentially an update on Xcode, Apple’s free integrated development environment (IDE) that will be powered by Anthropic’s Claude Sonnet model, according to reports. Apple’s plan is said to be introducing the software internally then deciding later whether to launch it publicly. Apple developed the proprietary Xcode suite of tools and began using it in 2003 to code, debug, test and vet submissions to the App Store for software that runs on all of its operating systems, from iOS to visionOS. Continue reading Apple Reportedly Using Anthropic’s Claude for AI Code Tool