Claude is a family of AI assistants developed by Anthropic. Named versions (Claude 3.5 Sonnet, Claude 3 Opus) are known for long context windows, coding ability, and safety-focused design.
StarryKit, a new open-source plugin from StarryKit, enables AI agents to create and edit visual designs such as slides, posters, and catalogs directly, with export to PPTX, PDF, SVG, PNG, JPEG, HTML, …
AutoGPT, the open-source project with over 180,000 stars, is embracing AI-generated pull requests rather than blocking them, according to founding AI engineer Nicholas Tindle. Tindle says the key is t…
In August 2026, AI-related content dominated every post on the blog, prompting the author to plan a return to non-AI topics such as childhood, education, fertility, housing, and dating. The roundup al…
A researcher from Anthropic reported a solution to the Hadamard matrix problem for order 668, crediting a team of three humans and Claude. The solution encodes Hadamard matrices for all previously unk…
Anthropic, the AI company behind the Claude chatbot, is adding watermarks to AI-generated text, images, and files worldwide to comply with the European Union AI Act's transparency code, which took eff…
Raycast released version 0.71 with a new Screen Awareness feature that lets Raycast AI access content from the focused app or window, eliminating the need to copy text or attach screenshots. The featu…
A developer created Bowls Above, an idle desktop game that rains balls while coding agents like Codex or Claude think, and stops when they yield, aiming to replace doomscrolling during coding waits. T…
Most production AI agents run on closed frontier models that developers cannot fine-tune, so continual learning must happen in the agent's environment rather than in model weights, according to a guid…
A million-token context window is still insufficient for complex, real-world agentic tasks, according to an analysis of LLM architecture and agent design. The author outlines six building blocks of co…
Anthropic announced it will embed invisible, machine-readable watermarks in text generated by its Claude chatbot, starting with new models launched in the EU on or after August 2, and will apply the m…
BCG projects that 50% to 55% of US jobs will be reshaped by AI within two to three years, while 10% to 15% could be eliminated over four to five years, with software engineering in the Amplified segme…
Samsung Electronics Co. reported that using Anthropic PBC's Claude AI model cut its semiconductor design and verification time from one month to two days, a 93% reduction, according to a Chosun Biz ar…
Google's Gemini AI model is losing market share to OpenAI's ChatGPT and Anthropic's Claude, according to new market data from Pangram, Similarweb, and OpenRouter. Pangram reports Gemini's share droppe…
A developer warns that AI coding agents are eliminating the middle class of software engineering by enabling junior developers to produce massive volumes of code they don't understand, overwhelming se…
Anthropic has added optional usage credits for paid Claude plans, allowing Pro, Max 5x, and Max 20x subscribers to continue using Claude after hitting included session limits at standard API rates, wi…
Anthropic's invisible watermarking of Claude outputs is causing technical friction for developers, who report that the statistical patterns degrade logical precision in long AI agent chains and introd…
Anthropic is shifting to an asset-light model by having strategic partners underwrite the construction of massive data center fleets, allowing the AI lab to scale its LLM agent capabilities without be…
Florian Herrengt warns that AI-assisted programming is creating unmanageable 'cognitive debt' as developers rely on tools like Claude to fix bugs they do not understand, leading to convoluted projects…
More than 75% of AI-first businesses took an average of 80 days longer to recover from security incidents than peers, and nearly half said AI was directly exploited in their most recent breach, accord…
A developer's evaluation of an agentic harness built around an LLM shows a 75% pass rate, averaging 1,088 tokens and $0.0033 per task across four tasks, including one adversarial task. The harness, no…