RobinsonLabs/ news

Anthropic Disables Flagship AI Models After US Export Ban

Anthropic launched Claude Fable 5 and Claude Mythos 5 on June 9, 2026, marketing them as the world's most powerful AI models. Just three days later, the Trump administration cited national-security concerns and issued an export-control directive barring any foreign national—whether inside or outside the United States, including foreign-national Anthropic employees—from accessing the models. Because Anthropic's architecture made it impractical to gate access on a per-user basis, the company abruptly disabled both Fable 5 and Mythos 5 for all customers as of June 12, 2026.

The move marks one of the most aggressive uses of export controls on generative AI to date and signals Washington's growing determination to treat cutting-edge models as strategic assets. Analysts say the precedent could affect other U.S. AI labs as the global compute and model race intensifies.

OpenAI Faces Sweeping Probe from 42 State Attorneys General

New York Attorney General Letitia James served the subpoena on behalf of a 42-state coalition, demanding records on OpenAI's advertising and user-engagement practices, how the company handles consumer and health data, interactions involving minors and seniors, model performance and internal policies, and a design flaw the subpoena calls 'sycophancy'—the tendency of ChatGPT to agree with users even when they are wrong. The investigation runs parallel to a separate lawsuit filed by Florida's attorney general alleging ChatGPT links to violent incidents.

OpenAI, which filed confidentially for a $1 trillion IPO just five days before the subpoena, said it would respond 'constructively' and already has measures in place to protect its customers. The multi-state probe adds to a growing wave of state-level scrutiny that could shape OpenAI's regulatory landscape as it prepares for its public debut.

Google Invests Up to $40 Billion in Anthropic in Landmark AI Bet

Announced on April 24, 2026, the deal deepens the partnership between Alphabet and Anthropic beyond Google's initial October 2025 TPU capacity agreement. Under the terms, Google Cloud will supply 5 gigawatts of dedicated compute capacity on Google's TPU infrastructure over five years, alongside chipmaker Broadcom for TPU-based capacity starting in 2027. The post-money valuation for Anthropic reached $350 billion, making it the largest single financial commitment to an AI startup outside Microsoft's OpenAI partnership.

The investment arrives as the AI industry's compute arms race intensifies—Amazon had expanded its own commitment to Anthropic to as much as $25 billion just days earlier. Google's dual role as both investor and compute provider gives it a strategic foothold in Claude's development while securing long-term demand for its custom TPU silicon.

ServiceNow and IBM Unveil Multi-Year Partnership to Modernize Legacy IT with AI

The partnership, announced June 11, 2026, brings IBM's Bob AI coding agent, Enterprise Application Runtime for Java, and watsonx.data data lakehouse onto the ServiceNow platform. ServiceNow customers will be able to scan and refactor legacy systems without starting from scratch, extend Workflow Data Fabric with watsonx.data for data quality and governance, and automate workflows using IBM's automation capabilities. The three-pillar approach targets two of the biggest barriers to enterprise AI adoption: the AI-ready data problem and the legacy application layer. The multi-year deal positions both companies to compete more aggressively in the enterprise AI modernization market.

ServiceNow Lays Off Hundreds of Employees, Citing AI Efficiencies

The layoffs were first reported on June 11, 2026, with a notable subset — 63 employees — specifically cut at the company's San Diego headquarters. According to Layoffs.fyi data, the restructuring was driven by AI productivity gains, with McDermott telling shareholders in April 2026 that the company expects to hold overall headcount flat into 2027 by using AI to offset natural attrition rather than backfilling roles. The San Diego site, where ServiceNow has a significant presence, saw 63 workers let go in the latest round. The company has been increasingly selective with hiring since its April 22, 2026 earnings call.

ServiceNow Warns of Active Exploitation of API Flaw Exposing Customer Instance Data

ServiceNow's security advisory said the flaw allowed unauthenticated users to access instance tables under certain circumstances, affecting a subset of hosted customer instances. The company applied a security update requiring authentication to the vulnerable endpoint. Security researchers later reported observing anomalous activity and customer data exposure in some instances, with one researcher confirming the flaw as active exploitation on June 9. The incident adds to a growing list of cloud service disruptions affecting enterprise ITSM platforms, and ServiceNow asked customers to review their instance configurations and audit access logs for signs of unauthorized queries.