The 10 biggest things in tech, Thu, Jun 25
The ten biggest stories in tech today, scored 1–10 and ranked from 150+ engineering sources. Free, updated every morning.
Critical zero-days under active exploitation
Three high-severity vulnerabilities are under active exploitation: a CVSS 9.8 code injection flaw in Lantronix EDS5000 devices (CVE-2025-67038) flagged by CISA with a federal remediation deadline, a Cisco Catalyst SD-WAN zero-day (CVE-2026-20245) exploited for two months before disclosure per Mandiant, and a 10M-install Chrome extension 'Adblock for YouTube' found to contain dormant arbitrary JavaScript injection capability.
#1 · security · scored 9/10
AI agent platforms ship broadly
Google (Gemini 3.5 Flash computer use), Vercel (AI SDK 7 with WorkflowAgent and MCP), Cloudflare (open-source Zero Trust agent skills), and Grab (Kubernetes-native Palana sandbox) all shipped production agent infrastructure this week, signaling a broad industry push to operationalize autonomous agents.
#2 · ai · scored 8/10
Google AI liability ruling in Germany
A German court held Google legally liable for false AI-generated search summaries, rejecting the defense that users should verify outputs independently and establishing that AI-generated content carries publisher-level responsibility — a significant precedent for AI liability across Europe.
#3 · security · scored 8/10
EU designates Amazon, Microsoft as cloud gatekeepers
The European Commission's preliminary DMA gatekeeper designation of Amazon and Microsoft imposes strict regulatory obligations that will reshape competitive dynamics across European cloud markets.
#4 · cloud · scored 8/10
IBM sub-nanometer chip breakthrough
IBM announced the world's first sub-1 nanometer chip technology using nanostack transistors, packing nearly 100 billion transistors and claiming scalability down to 1 Angstrom — a landmark milestone in semiconductor fabrication with major implications for future compute performance and energy efficiency.
#5 · infrastructure · scored 8/10
Europe tightens Big Tech regulation
The European Commission designated Amazon and Microsoft as cloud gatekeepers under the DMA, and a German court held Google liable for false AI-generated search summaries, collectively signaling a major European regulatory tightening on AI and cloud platforms.
#6 · industry · scored 8/10
OpenAI enters custom silicon
OpenAI's Jalapeño inference accelerator chip, co-developed with Broadcom and manufactured by Celestica, marks the company's first move into custom silicon to cut inference costs and reduce dependence on Nvidia.
#7 · ai · scored 7/10
AI-assisted cybercrime disruption
Microsoft used AI-powered link analysis to tie StealC and Amadey malware operations together, enabling shutdown of 200+ C2 servers in a racketeering suit, while Operation Endgame simultaneously dismantled two additional widely-used cybercrime tools in a coordinated global law enforcement action.
#8 · security · scored 7/10
AI agents and LLM security risks
Grab open-sourced Palana, a Kubernetes-native isolation platform for autonomous AI agents, as researchers simultaneously demonstrated that LLMs are structurally vulnerable to prompt injection due to role-block recognition rather than semantic understanding, and a novel Rust-based macOS malware called Gaslight weaponizes prompt injection to sabotage AI-assisted malware analysis tools.
#9 · security · scored 7/10
Software supply chain compliance pressure
The EU Cyber Resilience Act's SBOM, vulnerability reporting, and container compliance mandates are reshaping software supply chains, while Chainguard shipped drop-in remediated Java libraries for legacy CVEs, Azul launched a free JVM vulnerability scanner, and researchers found 6 new CVEs in the widely-deployed Curl library — all converging on a single narrative of escalating supply-chain security obligations.
#10 · security · scored 7/10