Welcome to AI Insider’s The Week Ahead in AI. See the key developments and events we’re watching June 28-July 4.
Weekend AI News Briefs
Google Caps Meta’s Gemini Use as AI Demand Strains Capacity
The Financial Times reported that Google has capped Meta’s access to its Gemini AI models after the social media company requested more computing capacity than Google could provide, delaying some internal AI projects and prompting Meta to encourage more efficient AI token usage. The restrictions highlight growing infrastructure constraints across the AI industry as demand for computing power outpaces available capacity, even among the largest technology companies. (The Financial Times)
AWS CEO Says Replacing Young Employees with AI is ‘One of the Dumbest Ideas’—And Bad for Business: ‘At Some Point the Whole Thing Explodes on Itself’
Amazon Web Services CEO Matt Garman rejected the idea that AI will eliminate entry-level software engineering jobs, arguing companies need junior talent to build future leaders and ideas, Fortune reported. Garman said Amazon plans to hire 11,000 interns and recent graduates in 2026 even as the company expands its use of AI, while acknowledging the technology will change many jobs rather than eliminate them. (Fortune)
How a ‘Coalition of the Willing’ Wants to Change the Agenda on AI
In a joint interview with Politico, former Commerce Secretary Gina Raimondo and former Indiana Gov. Eric Holcomb said their new nonprofit, RAISE US, will work with states and major technology companies to help workers prepare for the AI economy. Raimondo said the U.S. cannot win the AI race with China if it leaves American workers behind, warning that doing so could fuel political instability and lead to restrictions on AI development. Holcomb pointed out that the organization will focus on delivering measurable workforce results quickly, emphasizing that workers need new skills and pathways into the jobs created by AI. (Politico)
Firmus to Build 170,000 GPU AI Factory Campus with Nvidia for Global AI-Natives
Firmus Technologies said it has signed a strategic compute partnership with Nvidia through 2034 centered on a 360-megawatt AI factory campus in Batam, Indonesia, to expand access to AI computing for startups, enterprises and software vendors. The agreement includes up to 170,000 Nvidia AI accelerators across the Grace-Blackwell, Vera-Rubin and Vera platforms through 2028, with Firmus expecting $25 billion to $30 billion in revenue from committed customer agreements during the partnership’s first six years. (Firmus)
South Korean President to Unveil Massive AI and Chip Investment Drive
Reuters reported that South Korea is set to unveil three major projects focused on semiconductors, AI data centers and physical AI, including a proposed semiconductor hub in the country’s southwest that local media say could attract more than USD$651 billion in investment over several years. The government is expected to announce support for power, water, land, infrastructure and workforce development, while Samsung, SK Group and other major companies are expected to outline investment plans as part of the initiative. (Reuters)
Ford’s AI Wasn’t Smart Enough to Replace Veteran Engineers — So It Hired 350 of Them to Fix Quality Control Issues
Ford said it relied too heavily on AI rather than experienced engineers in vehicle development, prompting the automaker over the past three years to hire, rehire or promote 350 technical specialists to improve product quality, according to Yahoo Finance. Ford said the change has helped strengthen engineering oversight, improve AI training data and contributed to the company achieving the top ranking among mainstream automakers in J.D. Power’s initial quality study for the first time in 16 years. (Yahoo Finance)
I Don’t Hide My Use of AI. I Want My Kids to See How I Use It to Make Life Easier
A technology public relations executive described in Business Insider how her family uses generative AI for work, education and travel, arguing that the technology encourages curiosity and critical thinking when used intentionally rather than replacing it. She said her children are learning to question AI-generated answers instead of accepting them at face value, while her husband has replaced social media use with AI-assisted problem-solving. (Yahoo Finance)
How AI Helped the FBI Investigate the White House Correspondents’ Dinner Attack
Axios reported that the FBI used Exterro’s AI-powered digital forensics platform during its investigation into the attempted assassination at this year’s White House Correspondents’ Dinner to help process large volumes of digital evidence. Exterro said its software enables investigators to search and organize data from seized devices and accounts using AI while keeping evidence on secure, on-premises systems, with investigators remaining responsible for reviewing evidence and making charging decisions. (Axios)
Research & Innovation
Researchers Look at Impact of AI Support Tool in Real-world Primary Care Trial
A Gates Foundation-funded study published in Nature Medicine found that a generative AI clinical support tool safely improved clinical decision-making but did not significantly improve short-term patient outcomes in a trial involving more than 9,600 patients across 16 primary care clinics in Kenya. Researchers reported no statistically significant difference in 14-day treatment failure between AI-assisted and standard care, although the AI system improved clinical documentation and treatment planning while reducing antibiotic-related costs. (AI Insider)
AI Policy & Governance
A Midlife Crisis? IP and the Internet After 40
The House Judiciary Subcommittee on Courts, Intellectual Property, Artificial Intelligence, and the Internet will hold a hearing Tuesday, June 30, examining how intellectual property protections have evolved over four decades of the internet, with a focus on piracy, counterfeiting and abusive digital replicas. Lawmakers will hear testimony from representatives of SAG-AFTRA, Amblin Entertainment, the Software and Information Industry Association, IP House and Columbia Law School as they consider issues affecting U.S. intellectual property policy and enforcement. (Committee on the Judiciary)
Upcoming Events
AI Engineer World’s Fair 2026
June 29-July 2, San Francisco, Calif., the world’s largest technical AI conference brings together AI engineers, founders, researchers and technology leaders to explore the latest advances in AI systems engineering, agentic systems, inference, world models and Physical AI. The event features more than 400 sessions across 12 parallel tracks, 474+ speakers, 100+ expo partners, hands-on workshops and dedicated programming on Robotics & World Models, including side events focused on Physical AI deployment and applications. (AI Engineer)
Check out AI Insider every day for the latest in artificial intelligence and robotics.
