Microsoft has backed Anthropic in a legal challenge against the Pentagon over an AI and cloud computing contract, alleging procurement irregularities (Bloomberg). The International Energy Agency recommended a record release of 400 million barrels of oil to stabilize prices amid the U.S.-Israeli conflict with Iran (IEA). The U.S. Nuclear Regulatory Commission has scrapped rules requiring new reactors to withstand intentional aircraft impacts, aiming to simplify nuclear plant construction (NRC).
The intensifying conflict between Israel and the United States against Iran dominates today's headlines, with the most severe airstrikes yet launched and rising U.S. casualties. Meanwhile, the tech industry continues its rapid AI integration across consumer and enterprise products, while grappling with the deepfake dilemma—rolling out both creation tools and detection systems in tandem.
The United States and Israel launched their most intense airstrikes against Iran on Tuesday, marking a significant escalation in the regional conflict (WIRED, multiple outlets). The U.S. military reports hitting 16 Iranian mine-laying ships, while up to 150 American troops have reportedly been wounded in the ongoing operations. The UAE has activated air defense systems as tensions spread beyond the immediate theater. Markets, which had initially anticipated a swift resolution under President Trump, closed mixed as the situation deteriorated. In a notable development, Ukraine has dispatched a specialist drone team and interceptors to Jordan at Washington's request to help protect U.S. military bases from Iranian Shahed drones—a cost-effective countermeasure to the expensive Patriot missiles currently used for interceptions.
Google is significantly expanding Gemini AI integration across its Workspace suite, introducing features that let users generate complete documents and spreadsheets from natural language prompts while pulling data from Gmail, Chat, and Drive (Google). New tools include a "Help me create" function in Docs and a "Match writing style" feature to maintain consistent tone in collaborative documents. Adobe has launched its AI assistant for Photoshop in beta across web and mobile, enabling complex edits like object removal through conversational prompts, with paid subscribers receiving unlimited AI generations through April 9 (Adobe). Zoom is also making a significant push, launching photorealistic AI avatars that mimic user expressions, a new AI-powered office suite with Docs, Slides, and Sheets, and expanded AI Companion 3.0 capabilities (Zoom).
YouTube is extending its AI likeness detection technology to a pilot group of government officials, political candidates, and journalists, allowing them to identify and request removal of unauthorized deepfake content (YouTube). The system, functioning similarly to Content ID, scans for faces and requires government ID verification from participants. While removal requests will be evaluated with exceptions for parody and political critique, YouTube is exploring future options including pre-upload blocks and potential monetization paths for AI-generated content. Separately, Zoom's announcement includes deepfake-detection technology alongside its avatar creation tools—an acknowledgment that the same companies building synthetic media must also police it.
Thinking Machines Lab, the AI startup founded by former OpenAI executive Mira Murati, has signed a multi-year deal with Nvidia to deploy at least one gigawatt of Vera Rubin compute systems beginning in 2027 (Nvidia). The agreement includes a strategic investment from Nvidia into the lab, now valued at over $12 billion. In legal tech, Legora has reached a $5.55 billion valuation following a $550 million Series D led by Accel, more than tripling its valuation since October 2025 (Accel). The platform now serves 800 law firms. Meta, meanwhile, has acquired Moltbook, a social network for autonomous AI agents, to integrate its team into Meta Superintelligence Labs (Meta). Amazon is taking a more cautious approach: following AWS outages linked to AI agent errors, the company now requires senior engineers to sign off on any AI-assisted code changes made by junior and mid-level staff (Amazon).
Kansas has invalidated trans people's driver's licenses, requiring IDs to reflect birth-assigned gender—a move experts warn will extend discrimination to the internet through age-verification laws now active in over half of U.S. states (WIRED). Tech advocates caution that automated identity verification systems, which often rely on biased technology, risk outing or excluding marginalized groups from essential online services. Separately, Google is adding a visible toggle to its Photos app allowing users to switch back to classic search instead of the AI-powered "Ask Photos" feature, responding to user complaints about latency and accuracy issues with the Gemini-powered tool (Google).
UN investigators have found that Russia's deportation of Ukrainian children constitutes a crime against humanity, while the war continues to drive technological innovation—including armed robots now deployed on the battlefield and electronic warfare affecting navigation apps. In the United States, a new report documents record-high Islamophobia in 2025, linked to the Trump administration's crackdowns on protests and immigration. Meta's oversight board is urging the company to improve policing of fake AI videos during crises, calling current methods inadequate.