Skip to content
Personal blog. Opinions are my own. Always refer to official documentation.
AI Pulse · Daily Digest

Thursday, June 25, 2026

15 stories from the Gen AI ecosystem · curated & summarised

★ Top Story Ars Technica

OpenAI and Broadcom Unveil "Jalapeño" — OpenAI's First Custom AI Inference Chip

OpenAI has revealed its first custom silicon, an ASIC called Jalapeño, co-developed with Broadcom and purpose-built for large language model inference at scale. The chip is designed to reduce OpenAI's dependence on third-party GPU suppliers like NVIDIA as demand for AI compute continues to surge. The announcement marks a significant step in OpenAI's vertical integration strategy and intensifies the broader race among AI labs to control their own hardware stack.

Read the original →
TechCrunch AI

AI Researchers Keep Leaving Google — This Time for Anthropic

Two more senior Google AI researchers, Jonas Adler and Alexander Pritzel, are departing for Anthropic, continuing a notable talent drain from the search giant. The exits follow high-profile departures including Noam Shazeer and Nobel Prize winner John Jumper, raising questions about Google's ability to retain top-tier AI talent. The trend underscores the intensifying competition for elite researchers as AI labs compete aggressively on compensation, autonomy, and research culture.

Read more →
TechCrunch AI

Europe Pushes Back Against U.S. Chip Export Controls

European governments and semiconductor industry leaders are mounting resistance to Washington's expanding chip export restrictions, particularly the proposed MATCH Act, which would extend controls to older-generation deep ultraviolet (DUV) lithography tools. ASML, the Dutch company that holds a near-monopoly on advanced chip-making equipment, has been vocal about the economic and geopolitical risks of such sweeping restrictions. The standoff highlights growing transatlantic friction over how far the U.S.-led tech decoupling from China should extend.

Read more →
TechCrunch AI

Cerebras Stock Plunges After First Post-IPO Earnings Report

AI chipmaker Cerebras saw its stock tumble sharply after its debut earnings report as a public company, in which it forecast a narrower gross margin in its core business than investors had anticipated. CEO Andrew Feldman pushed back, arguing the market misread the guidance, but the damage was done as investors reacted to the compressed profitability outlook. The selloff raises broader questions about valuation sustainability for AI hardware companies amid fierce competition from NVIDIA and new entrants.

Read more →
TechCrunch AI

AI Was Supposed to Kill Engineering Jobs — New Data Says the Opposite

Despite widespread predictions that AI would hollow out software engineering roles, new data from venture firm SignalFire shows engineers are actually accounting for a growing share of new hires across the tech industry. The findings suggest that AI tools are augmenting engineering productivity rather than replacing headcount, and that demand for engineers who can build and deploy AI systems is accelerating. The data challenges the dominant layoff narrative and points to a more nuanced labor market reshaping underway.

Read more →
TechCrunch AI

Agility Robotics Plans $2.5B SPAC Listing

Humanoid robotics company Agility Robotics, maker of the bipedal Digit robot, has announced plans to go public through a SPAC merger at a $2.5 billion valuation, expecting to raise approximately $620 million in proceeds. The Oregon State University spinout has been deploying Digit in warehouse environments and counts Amazon among its early commercial partners. The listing would make Agility one of the first pure-play humanoid robotics companies to reach public markets, a significant milestone for the sector.

Read more →
TechCrunch AI

The AI Token Budget Wars: Companies Scramble to Ration Employee AI Usage

Enterprises are increasingly implementing token caps and usage governance policies after discovering employees are burning through AI compute budgets on routine or trivial tasks. The phenomenon, dubbed "tokenmaxxing," has prompted IT and finance teams to introduce rationing systems, tiered access, and cost-allocation frameworks for AI tools. The trend reveals a maturing phase of enterprise AI adoption where the focus is shifting from access and experimentation to cost discipline and ROI accountability.

Read more →
The Verge AI

The $27 Million AI Super PAC Proxy War Ends in a Draw

A closely watched New York congressional primary that became a proxy battle between AI industry interests ended inconclusively, with AI-friendly candidate Alex Bores narrowly losing the Democratic primary despite — or perhaps because of — $27 million in outside spending from pro-AI super PACs linked to Anthropic and OpenAI backers. Bores' opponent benefited from a backlash against the heavy-handed corporate intervention in a local race. The outcome signals that AI companies' attempts to directly shape U.S. legislative representation face significant public resistance.

Read more →
TechCrunch AI

Anthropic's Claude Tag Embeds Always-On AI Into Slack Workflows

Anthropic has launched Claude Tag, a feature that integrates Claude as a persistent AI teammate within Slack, allowing it to learn organizational context, institutional knowledge, and workflows over time. Beyond productivity gains, the product is a strategic move to embed Claude deeply into enterprise communication infrastructure, making it harder for companies to switch to rival AI providers. The launch intensifies competition with Microsoft Copilot and Google Workspace AI for dominance in the enterprise collaboration layer.

Read more →
Ars Technica

Oracle's 21,000 Layoffs Fund Debt-Driven AI Infrastructure Buildout

Oracle has cut 21,000 employees as part of a sweeping restructuring designed to redirect capital toward massive data center and AI infrastructure investments, financed largely through debt. The company is betting that its cloud and AI infrastructure buildout will position it as a major player in the enterprise AI stack, competing with AWS, Azure, and Google Cloud. The scale of the workforce reduction underscores how aggressively legacy enterprise tech companies are pivoting their cost structures to fund AI ambitions.

Read more →
MIT Tech Review

Stripe, Anthropic, and OpenAI Back New Initiative to Defeat the Common Cold

Stripe, co-founded by Patrick and John Collison, is leading a new funding initiative alongside Anthropic and OpenAI to support research aimed at preventing common respiratory infections, including the common cold, for which there is currently no vaccine or reliable preventive treatment. The effort reflects a growing trend of AI-era tech wealth being directed toward ambitious biomedical moonshots. It also signals the expanding ambitions of leading AI companies beyond software, into life sciences and public health.

Read more →
TechCrunch AI

Figma Overhauls Platform With AI Motion Tools, Shaders, and Full-Stack Dev Features

At its annual Config conference, Figma unveiled a major platform update adding a new code layer, AI-powered motion graphics and shader tools, and a reimagined canvas optimized for full-stack development. The update positions Figma as more than a design tool, pushing into the territory of AI-assisted development environments where design and code converge. The additions put Figma in more direct competition with tools like Cursor and GitHub Copilot for the attention of developers and designers working on AI-native products.

Read more →
The Verge AI

Hollywood Studios Are Shunning a Biopic About Sam Altman

Several major Hollywood studios and distributors — including Netflix, A24, Focus Features, and Warner Bros.' Clockwork — have reportedly passed on acquiring distribution rights to "Artificial," director Luca Guadagnino's biographical drama about OpenAI CEO Sam Altman. The widespread reluctance from studios with significant AI partnerships or licensing deals raises pointed questions about whether financial entanglements with OpenAI are influencing editorial and acquisition decisions. Only Neon and Mubi remain interested, making the film's path to wide release uncertain.

Read more →
The Verge AI

Midjourney's Pivot to Medical Body Scanners Draws Skepticism

Midjourney, best known as an AI image generator, surprised the industry by announcing a pivot into medical imaging with a futuristic water-immersion ultrasound scanner it claims could rival MRI quality. Critics and medical experts have raised serious concerns about the lack of clinical evidence supporting the device's capabilities, and the announcement has been met with widespread skepticism about whether an image-generation company has the expertise to deliver on such claims. The move highlights the growing tension between AI startup ambition and the rigorous validation standards required in healthcare.

Read more →
NVIDIA Blog

NVIDIA Powers 81% of the World's 500 Fastest Supercomputers

The latest TOP500 supercomputer rankings, released at the ISC High Performance conference in Hamburg, show NVIDIA technology powering more than 400 of the world's 500 fastest systems — an 81% market share that cements the company's dominance in high-performance and AI computing infrastructure. The figures come as rivals including AMD, Intel, and custom silicon from hyperscalers attempt to chip away at NVIDIA's grip on the market. The data reinforces why NVIDIA remains the central hardware dependency for both scientific computing and AI model training at scale.

Read more →
No stories from this source today.

Summaries are AI-generated. Click through to read the original reporting.

Summaries are AI-generated. Click through to read the original reporting.