Model Launch

Twitter

Facebook

Copy Link

Top News

White House drafts guidance to restore Anthropic access as Pentagon legal fight continues

The White House is developing guidance that could allow federal agencies to bypass Anthropic's supply chain risk designation and access new AI models including Mythos. The move signals a potential reversal after the Trump administration previously blacklisted the company over its refusal to remove guardrails against autonomous weapons and mass domestic surveillance.

Policy8 sources

White House drafts guidance to restore Anthropic access as Pentagon legal fight continues

Policy8 sources

OpenAI releases Privacy Filter to scrub personal data before it reaches AI models

OpenAI launched Privacy Filter, a free open-source tool that detects and removes personally identifiable information from text before it reaches cloud servers. The 1.5-billion-parameter model runs on standard laptops, achieving 96% accuracy on industry benchmarks. Released under Apache 2.0 license, it addresses growing concerns about sensitive data exposure in AI workflows.

Technology2 sources

OpenAI releases Privacy Filter to scrub personal data before it reaches AI models

Technology2 sources

Sam Altman accuses Anthropic of fear-based marketing over Claude Mythos cybersecurity model

OpenAI CEO Sam Altman criticized Anthropic's marketing strategy for Claude Mythos, calling it fear-based tactics to keep AI in elite hands. The accusation highlights deepening tensions between the AI giants as they compete for enterprise clients and prepare for public offerings, with Anthropic's revenue hitting $30 billion annually.

Technology5 sources

Sam Altman accuses Anthropic of fear-based marketing over Claude Mythos cybersecurity model

Technology5 sources

Moonshot AI's Kimi K2.6 coordinates 1,000 agent swarms running for days on complex tasks

Moonshot AI unveiled Kimi K2.6, an open-source AI model that orchestrates up to 1,000 collaborating agents to handle complex tasks autonomously. The model ran continuously for five days in one demonstration, managing monitoring and incident response without human intervention. But the release exposes critical gaps in enterprise orchestration frameworks not designed for such long-running, stateful execution.

Technology2 sources

Moonshot AI's Kimi K2.6 coordinates 1,000 agent swarms running for days on complex tasks

Technology2 sources

OpenAI launches ChatGPT Images 2.0 with text rendering that finally works

OpenAI released ChatGPT Images 2.0, a new AI image model that can accurately generate text in multiple languages and create complex visual content like infographics and marketing materials. The model includes thinking capabilities, web information access, and can generate up to eight images from a single prompt while maintaining visual consistency.

Technology28 sources

OpenAI launches ChatGPT Images 2.0 with text rendering that finally works

Technology28 sources

OpenAI releases GPT-5.5, advancing toward AI superapp with enhanced coding and research tools

OpenAI unveiled GPT-5.5 on Thursday, calling it its smartest and most intuitive AI model yet. Released just seven weeks after GPT-5.4, the model brings significant improvements in agentic coding, knowledge work, and scientific research. President Greg Brockman says it's a step toward OpenAI's planned superapp that will unify ChatGPT, Codex, and AI browser capabilities.

Technology35 sources

OpenAI releases GPT-5.5, advancing toward AI superapp with enhanced coding and research tools

Technology35 sources

Mike Krieger exits Figma board as Anthropic enters web design space with competing tools

Mike Krieger, Anthropic Chief Product Officer, resigned from Figma's board on April 14, the same day reports emerged that Anthropic Opus 4.7 will include design capabilities competing with Figma's core products. The move highlights growing tensions as AI labs expand into software businesses, fueling SAASpocalypse fears among investors.

Business2 sources

Mike Krieger exits Figma board as Anthropic enters web design space with competing tools

Business2 sources

OpenAI unveils GPT-Rosalind AI model to accelerate drug discovery and life sciences research

OpenAI released GPT-Rosalind, its first domain-specific AI model built for life sciences research. Named after DNA pioneer Rosalind Franklin, the model aims to compress the 10-to-15-year drug development timeline by helping scientists synthesize evidence and generate biological hypotheses. Access is restricted through a trusted access program to enterprise partners including Amgen, Moderna, and Thermo Fisher Scientific.

Technology14 sources

OpenAI unveils GPT-Rosalind AI model to accelerate drug discovery and life sciences research

Technology14 sources

Google's Flash TTS gives developers director-level control over AI voices in 70+ languages

Google DeepMind launched Gemini 3.1 Flash TTS, a text-to-speech model that lets users direct vocal style, delivery, and pace through natural language commands. Scoring 1,211 on the Artificial Analysis TTS leaderboard, the model supports over 70 languages with regional accents and includes SynthID watermarks for detecting AI-generated content.

Technology2 sources

Google's Flash TTS gives developers director-level control over AI voices in 70+ languages

Technology2 sources

Google DeepMind unveils Gemini Robotics-ER 1.6 as robot dogs gain precision instrument reading

Google DeepMind launched Gemini Robotics-ER 1.6, a foundation AI model for robotics that enables robot dogs like Boston Dynamics' Spot to read analog instruments with 98% accuracy. The model introduces agentic vision combining visual reasoning with code execution, allowing robots to interpret complex gauges, thermometers, and pressure readings in industrial environments. This marks a significant leap from the 23% accuracy of its predecessor.

Technology2 sources

Google DeepMind unveils Gemini Robotics-ER 1.6 as robot dogs gain precision instrument reading

Technology2 sources

Anthropic releases Claude Opus 4.7, outperforming GPT-5.4 on coding benchmarks

Anthropic has launched Claude Opus 4.7, its most capable publicly available AI model, with benchmark-leading scores on software engineering tasks. The large language model achieves 64.3% on SWE-bench Pro, surpassing OpenAI's GPT-5.4 at 57.7% and Google's Gemini 3.1 Pro at 54.2%. While positioned as less capable than the restricted Claude Mythos Preview, Opus 4.7 introduces new cybersecurity safeguards and delivers significant improvements in coding, vision, and multi-step reasoning.

Technology26 sources

Anthropic releases Claude Opus 4.7, outperforming GPT-5.4 on coding benchmarks

Technology26 sources

Anthropic's Mythos AI model triggers global alarm over cybersecurity and banking threats

Anthropic's new Mythos AI model has sparked urgent concerns among governments and financial officials worldwide over its ability to detect and exploit software vulnerabilities faster than current cybersecurity defenses can respond. The model has already uncovered thousands of high-severity flaws in major operating systems and web browsers, prompting emergency meetings between US Treasury officials, Federal Reserve leaders, and major banks to assess the risks to the global banking system.

Policy39 sources

Anthropic's Mythos AI model triggers global alarm over cybersecurity and banking threats

Policy39 sources

Microsoft launches MAI-Image-2-Efficient, a 41% cheaper and 22% faster AI image model

Microsoft unveiled MAI-Image-2-Efficient, a cost-effective image generation model priced at $19.50 per million image output tokens—41% cheaper than its flagship. The faster AI image model delivers 4x high throughput efficiency per GPU and ships just weeks after the original, marking the company's rapid push to build an independent AI stack beyond OpenAI.

Technology3 sources

Microsoft launches MAI-Image-2-Efficient, a 41% cheaper and 22% faster AI image model

Technology3 sources

Anthropic's Mythos AI model sparks global concern over advanced hacking capabilities

Anthropic restricted public release of its Mythos AI model after it demonstrated unprecedented cybersecurity skills, completing complex 32-step attacks and finding vulnerabilities in every major operating system. The UK's AI Security Institute confirmed the threat is real, while regulators and banks scramble to assess risks and demand equal access to defensive tools.

Technology22 sources

Anthropic's Mythos AI model sparks global concern over advanced hacking capabilities

Technology22 sources

OpenAI launches GPT-5.4-Cyber model days after Anthropic's Mythos sparks global security concerns

OpenAI unveiled GPT-5.4-Cyber, a specialized cybersecurity model, just days after Anthropic released its Mythos model with limited access due to security concerns. Both AI systems can detect software vulnerabilities at unprecedented speed, prompting urgent meetings between US Treasury officials and Wall Street leaders about potential risks to critical infrastructure.

Technology30 sources

OpenAI launches GPT-5.4-Cyber model days after Anthropic's Mythos sparks global security concerns

Technology30 sources

Prev 1…456…10 Next