The afternoon briefing.
Top intelligence agencies warn of imminent AI cyber catastrophes, Google DeepMind invests in AI filmmaking, and Valve prices its new Steam Machine.
AI Cyber Threats. Top intelligence agencies warn of imminent AI-driven cyber catastrophes, stating the timeline is months, not years. Breaches are expected to occur as AI capabilities are increasingly deployed in both offensive and defensive cyber operations. This rapid acceleration of AI in cyber warfare presents a critical challenge for organizational leaders globally. The urgency highlights the need for immediate and robust security measures.
AI Investment & Strategy. Google DeepMind is investing $75 million in A24 to develop AI filmmaking tools, marking Google's first stake in a film studio. Meanwhile, Anthropic is partnering with Micron to co-design AI memory architecture, with Micron also investing in Anthropic's Series H round. These moves highlight a broader trend of AI labs expanding beyond model provision to offer integrated services and secure critical hardware supply. Anthropic is also transforming enterprise collaboration with its new Artifacts feature.
AI Policy & Ethics. The New York City House primary has become a key battleground in an "AI civil war," with AI-focused Super PACs spending heavily to influence early legislation. Separately, Anthropic's Claude chatbot may soon require users to verify their identity and age in certain circumstances, citing privacy policy changes. These developments underscore growing scrutiny and the complex ethical and political dimensions emerging alongside AI's rapid advancement.
Hardware & Infrastructure. Valve has priced its new Steam Machine at $1,049, with shipments beginning June 29th, though availability will be limited by a randomized purchase queue. The company also announced that users will be able to build their own Steam Machine using SteamOS for desktop starting with the 3.8 release. This comes as concerns grow over data center locations, with almost 80% now constructed in natural disaster zones, exacerbating climate risks.
SpaceX inks compute deal with Reflection AI, an open source AI lab
Reflection AI will pay $150 million a month for immediate access to Nvidia's latest GB300 AI chips and supporting hardware across SpaceX's Colossus 2 data center. This multi-year deal, starting July 1, 2026, aims to bolster open-source AI development.
Top Intel Agencies Say AI-Driven Cyber Catastrophes Are Imminent
A coalition of top intelligence agencies warns that AI-driven cyber breaches are imminent, with a timeline of months, not years. They advise organizational leaders to prepare for significant cyber catastrophes.
Tata Electronics, a major tech supplier to Apple and Tesla, confirms data breach
Tata Electronics, a key supplier to Apple and Tesla, has confirmed a data breach. The incident occurs as the company expands its role in global technology supply chains.
New unpatchable flaw in Apple chips opens the door to an iPhone jailbreak
European cybersecurity company Paradigm Shift has detailed an unpatchable flaw in Apple chips, which could allow hackers to unlock and break into older iPhones. This exploit opens the door for new jailbreaking techniques.
Google DeepMind bets $75M on AI’s future in Hollywood with A24 deal
Google DeepMind and A24 are collaborating to build AI filmmaking tools, with Google investing $75 million in the film studio. This partnership aims to expand storytelling possibilities for future filmmakers.
Anthropic and Micron want to co-design AI memory architecture
Micron is investing in Anthropic's Series H round and secured a multi-year deal to supply memory for Claude's infrastructure. Anthropic co-founder Tom Brown emphasizes memory's critical role in training and running Claude models.
Anthropic Aims to Transform Enterprise Collaboration With Artifacts
Anthropic is launching Artifacts to transform enterprise collaboration, highlighting a trend among AI labs to offer services beyond just models. This move positions Anthropic more like a cloud provider.
Nvidia gets all agentic about supercomputing for scientific research
Nvidia suggests that tireless AI agents could help scientists conduct research that humans alone cannot achieve. The GPU giant is exploring the application of agentic AI in supercomputing for scientific discovery.
New lightweight, self-propagating crypto stealing malware delivered by USB spotted by Microsoft
Microsoft researchers have identified a new worm-like infostealer called Crypto Clipper, which is delivered via USB and hunts for vulnerable cryptocurrency wallets. This malware is designed to steal crypto assets.
Google makes Interactions API the default interface for Gemini models and agents
Google DeepMind has designated the Interactions API as the default interface for Gemini models and agents, replacing the older generateContent API. New agent features will exclusively ship through this simplified schema.
New York City House primary emerges as key battleground in ‘AI civil war’
AI-focused Super PACs are spending heavily in the 2026 midterms, with half of all spending concentrated on a single Manhattan congressional race. This highlights the industry's push to influence early AI legislation.
Valve's Steam Machine ships June 29 for $1,049, but you probably won't be able to buy one yet
Valve's Steam Machine is set to ship on June 29 for $1,049, utilizing a randomized purchase queue to ensure fairness. The company also plans to allow users to build their own Steam Machines with SteamOS for desktop.
NHTSA investigating alleged Tesla Autopilot crash that killed woman in her home
The NHTSA is investigating a fatal crash where a Tesla driver, allegedly using Autopilot, crashed into a woman's home, killing her. This incident raises further questions about the safety of automated driver-assistance systems.
Almost 80% of data centers constructed in natural disaster zones
New data reveals that nearly 80% of data centers are built in natural disaster zones, making them vulnerable to climate risks. This trend exacerbates environmental concerns while also putting critical infrastructure at risk.