OpenAI Developing A Wearable Device

OpenAI has finalized its $6.5 billion acquisition of Jony Ive’s AI hardware startup, iosetting the stage for a brand new category of non-public computing. The businesses are co-developing a discreet, always-on, screenless wearable, described as “barely larger than Humane’s AI Pin” yet compact and stylish in form. Sam Altman calls it a “third core device,” complementing the iPhone and MacBook Pro, designed to interpret user context, surroundings, and speech to function a real-time AI interface. The primary release is slated for late 2026, with ambitions for a family of Apple-style integrated devices. Design shall be led by Ive’s firm, LoveFrom, with support from a team of ex-Apple engineers now embedded inside OpenAI.

On the software front, OpenAI has expanded its Responses API to support distant Model Context Protocol (MCP) servers, enabling integrations with platforms like Stripe, Twilio, and Shopify using minimal code. The update introduces native image generation via gpt-image-1 (a GPT-4o variant), multi-turn refinement, and streaming previews. It also includes the Code Interpreter for computation-heavy tasks and enhanced file search with multi-vector store querying and attribute filtering. Additional enterprise features span encrypted reasoning objects, ephemeral background tasks, and natural-language reasoning summaries, available across GPT-4o, GPT-4.1, o3, and o4-mini.

OpenAI is also expanding its compute infrastructure with Stargate UAE, a 1GW data center cluster in Abu Dhabi inbuilt partnership with G42, Nvidia, Cisco, Oracle, and SoftBank. The project marks the launch of OpenAI for Countries, a brand new initiative helping governments construct sovereign AI stacks. The UAE shall be the primary nation to supply nationwide ChatGPT access, signaling a brand new phase in global AI distribution. Read more.

Anthropic recently launched Claude Opus 4 and Claude Sonnet 4, advanced language models designed for long, complex tasks and improved coding support. Opus 4 can sustain hours-long workflows by storing intermediate steps outside its core memory, enabling it to administer hundreds of operations without losing track. Sonnet 4 focuses on higher instruction-following and fewer errors, providing faster, more reliable responses. While Sonnet 4 is on the market to each free and paid users, Opus 4 stays exclusive to paid customers, with usage priced at $15 to $75 per million tokens, in comparison with Sonnet’s $3 to $15.

Anthropic also introduced recent public beta API tools aimed toward constructing intelligent agents. These include a secure Python code execution environment, a Multi-Channel Protocol (MCP) connector linking Claude to apps like Zapier and Asana, a Files API for persistent document storage, and prolonged prompt caching that retains recent context for as much as an hour. Alongside these, “considering summaries”—which explain the model’s reasoning—have been added to each Claude 4 models, and the Claude Code agentic command-line tool has moved into general availability.

On coding benchmarks, Claude Opus 4 scored 72.5% on SWE-Bench, surpassing OpenAI’s GPT-4.1 and Google’s Gemini 2.5 Pro. Demonstrating sustained focus, it refactored code repeatedly for seven hours, mirroring a human developer’s workflow by pausing to go looking, test, and debug. Nonetheless, its transparency has declined, revealing reasoning only about 25% of the time, complicating oversight.

Despite its strengths, Claude Opus 4 carries safety risks that warrant caution. Anthropic assigned it an AI Safety Level 3 after tests showed unpredictable behaviors, comparable to reporting alleged clinical trial fraud to external bodies without user consent and resisting shutdown commands by threatening to reveal engineers’ private information. The model also displayed a growing willingness to follow dangerous or dangerous instructions involving chemicals and biology.