Wiredframe Radar logo

Category intelligence

AI News Briefing — June 25, 2026

55 current items analyzed and ranked.

Executive synthesis

AI News Summary

Compute infrastructure dominated the day. The move signals frontier labs vertically integrating to control inference economics.

Mistral shipped OCR 4 (claimed wins in 72% of blind tests), and Gradium launched speech-translation models it says beat gpt-realtime-translate.

Key Themes

AI Hardware and Compute Infrastructure · 10Inference Efficiency and Research · 2Agentic AI and Workplace Agents · 7AI Policy and Politics · 9AI Talent Movement · 1Model and Capability Releases · 5US-China AI Competition · 2AI Labor, Privacy and Society · 6

Primary evidence

Top Ranked Signals

News Ars Technica - All content Jun 24

OpenAI and Broadcom announce chip designed for LLM inference at scale

By Samuel Axon

80 score
AI Analysis

OpenAI and Broadcom announced Jalapeno, a custom ASIC designed specifically for large language model inference in data centers. Both companies frame it as the first generation of a long-term silicon roadmap aimed at running models at scale.

OpenAI, the company behind ChatGPT and Codex and the models those tools utilize, and Broadcom, an established silicon supplier, have announced a new chip called Jalapeño, designed specifically for large language model inference in data centers. The chip is intended to be deployed at large data centers, both companies claim this is just the first generation in a long-term project that will see chips refined over time.Read full article Comments
AI HardwareCompute InfrastructureOpenAI
News AI News & Artificial Intelligence | TechCrunch Jun 24

OpenAI unveils its first custom chip, built by Broadcom

By Russell Brandom

78 score
AI Analysis

TechCrunch reports OpenAI unveiled its first custom processor, Jalapeno, built with Broadcom and tailored for the specific demands of OpenAI inference systems. It marks OpenAI entry into bespoke AI silicon.

Named Jalapeño, the new processor was designed specifically for the unique needs of OpenAI's inference systems.
AI HardwareOpenAICompute Infrastructure
News OpenAI News Jun 24

OpenAI and Broadcom unveil LLM-optimized inference chip

By Unknown

78 score
AI Analysis

OpenAI official announcement describes Jalapeno, a custom AI chip built with Broadcom and optimized for LLM inference to improve performance, efficiency, and scale. It represents OpenAI move into bespoke silicon.

OpenAI and Broadcom introduce Jalapeño, a custom AI chip built for LLM inference to improve performance, efficiency, and scale across AI systems.
AI HardwareOpenAICompute Infrastructure
News The Decoder Jun 24

OpenAI and Broadcom unveil "Jalapeño," a custom chip built for LLM inference

By Maximilian Schreiner

76 score
AI Analysis

The Decoder reports OpenAI and Broadcom unveiled the Jalapeno chip tailored for LLM inference, with plans to run it at scale by late 2026. It marks OpenAI adding custom hardware to its stack.

OpenAI is adding custom hardware to its tech stack. The "Jalapeño" chip, developed with Broadcom, is tailored for large language model inference and is set to run at scale by late 2026. The article OpenAI and Broadcom unveil "Jalapeño," a custom chip built for LLM inference appeared first on The Decoder.
AI HardwareOpenAICompute Infrastructure
News Feed: Artificial Intelligence Latest Jun 24

Qualcomm Buys Buzzy Chip Startup Modular for Nearly $4 Billion

By Lauren Goode

70 score
AI Analysis

Qualcomm agreed to acquire AI chip software startup Modular for nearly 4 billion dollars, expanding from edge devices into data center infrastructure. Modular built one of the most prominent AI compiler and runtime stacks of the era.

Modular, one of the most promising chip software startups of the AI era, heads for a multibillion-dollar exit.
AI HardwareAcquisitionsCompute Infrastructure
News aibusiness Jun 24

Qualcomm to Acquire AI Platform Developer Modular

By Scarlett Evans

68 score
AI Analysis

Qualcomm announced it will acquire AI platform developer Modular, expanding its AI infrastructure ambitions from edge devices into data centers. The deal deepens Qualcomm push into the data-center AI stack.

The move expands the chipmaker’s AI infrastructure ambitions from edge devices to data centers.
AI HardwareAcquisitionsCompute Infrastructure
News aibusiness Jun 24

Blackstone Commits $30B as Japan’s AI Battle Heats Up

By Graham Hope

66 score
AI Analysis

Blackstone committed 30 billion dollars to AI data center development in Japan as competition for AI infrastructure in the country intensifies. The plan joins a wave of multibillion-dollar data center deals.

The plan joins a host of other multi-billion-dollar AI data center deals.
Compute InfrastructureData CentersInvestment
News AI News & Artificial Intelligence | TechCrunch Jun 24

Agility Robotics plans to go public via SPAC in a $2.5B deal

By Kirsten Korosec

58 score
AI Analysis

Humanoid robotics startup Agility Robotics plans to go public via a SPAC in a 2.5 billion dollar deal expected to raise about 620 million dollars. The Oregon State spinout is targeting commercial scaling.

Agility Robotics, the humanoid robotics startup that spun out of Oregon State University in 2015, expects to generate $620 million in proceeds.
RoboticsFundingEmbodied AI
News The Decoder Jun 24

Snowflake CEO finds GLM-5.2 competitive with Opus 4.7 at a fraction of the cost

By Matthias Bastian

58 score
AI Analysis

Snowflake CEO reported that Zhipu AI GLM-5.2 nearly matched Claude Opus 4.7 on a 103-task coding benchmark at roughly one-fifth the cost per output token, though using nearly twice as many tokens. The result intensifies price pressure on Western labs.

Zhipu AI's GLM-5.2 nearly matches Claude Opus 4.7 in a Snowflake benchmark with 103 coding tasks at one-fifth the cost per output token. But the Chinese model burns through nearly twice as many tokens per task. Still, that pricing gap is putting real pressure on Anthropic and OpenAI, and could rattle the valuations of Western AI labs. The article Snowflake CEO finds GLM-5.2 competitive with Opus 4.7 at a fraction of the cost appeared first on The Decoder.
US-China CompetitionModel BenchmarksAI Economics
News The Decoder Jun 24

Claude Tag embeds Anthropic's AI in Slack, already writes 65 percent of internal code, company says

By Maximilian Schreiner

58 score
AI Analysis

Continuing our coverage of the Claude Tag launch from yesterday, Anthropic launched Claude Tag, letting teams summon its AI in any Slack channel by tagging it and assigning tasks. The company says the tool already generates 65 percent of code on its product team.

Claude Tag lets teams bring Anthropic's AI into Slack by tagging @Claude in any channel and assigning it tasks. Internally, the tool already generates 65 percent of the code on Anthropic's product team, the company says. The article Claude Tag embeds Anthropic's AI in Slack, already writes 65 percent of internal code, company says appeared first on The Decoder.
Agentic AIAnthropicAI at Work
News AI | The Verge Jun 24 Old anchor

OpenAI reveals its first AI processor: Jalapeño

By Emma Roth

55 score
AI Analysis

The Verge details OpenAI new Jalapeno ASIC, built with Broadcom for AI inference rather than training, arriving about nine months after the partnership was first revealed. The chip targets running ChatGPT and Codex agents at scale.

OpenAI has just revealed a new "intelligence processor" chip for AI servers made in partnership with Broadcom. The chip, called Jalapeño, is designed to power current and future large language models, according to an announcement on Wednesday. Jalapeño is an ASIC (Application-Specific Integrated Circuit), meaning it's designed for a specific purpose: AI inference. With AI inference, models process a user's request to run an agent like Codex or offer a response from ChatGPT
AI HardwareOpenAICompute Infrastructure
News aibusiness Jun 24 Old anchor

OpenAI and Broadcom Introduce AI Inference Chip

By Esther Shittu

55 score
AI Analysis

Coverage notes OpenAI and Broadcom inference chip could let model makers offer lower token prices, potentially easing enterprise cost concerns. It frames Jalapeno as a lever on AI economics.

The chip would give AI model makers the option to offer lower token prices, possibly easing the concerns of businesses worried about higher token costs.
AI HardwareAI EconomicsOpenAI