Tom's Hardware reported on March 15, 2026 that Markus 'Doom' Gaasedelen unveiled the Xbox One 'Bliss' exploit at RE//verse 2026. The report says the double voltage glitching attack reaches the boot ROM path, enables unsigned code at every level, and is effectively unpatchable because it targets silicon.
On March 11, 2026, Meta launched new anti-scam tools across WhatsApp, Facebook, and Messenger, including device-linking warnings, suspicious friend-request alerts, and broader AI scam review. Meta also said it removed more than 159 million scam ads last year and took down 10.9 million accounts linked to scam centers.
Anthropic said it detected industrial-scale campaigns by DeepSeek, Moonshot, and MiniMax to extract Claude outputs at scale. The company said the activity involved more than 16 million exchanges through about 24,000 fraudulent accounts and that it is investing in detection and response tooling.
GitHub said AI coding agents can now invoke secret scanning through the GitHub MCP Server before a commit or pull request. The feature is in public preview for repositories with GitHub Secret Protection enabled.
GitHub on 2026-03-09 detailed how Agentic Workflows are secured on top of GitHub Actions. The article is significant because it treats agents as untrusted components, isolates them from secrets, and stages writes before they can affect a repository.
Kotaku reports that Duet Night Abyss acknowledged a March 18, 2026 launcher update that spread malware, promising security improvements and in-game compensation after calling the breach a serious wake-up call.
Anthropic said on March 17, 2026 that open source security is becoming more important as AI grows more capable. In its X post, the company said it is donating to the Linux Foundation to help secure the software foundations AI depends on.
Meta said on March 11, 2026 that it is expanding anti-scam features across WhatsApp, Facebook, and Messenger while using more AI to detect celebrity, public-figure, and brand impersonation. The company also said it will expand advertiser verification so verified advertisers account for 90% of ad revenue by the end of 2026, up from 70% today, and disclosed that it removed more than 159 million scam ads in 2025.
Perplexity said on March 17, 2026 that it is launching Comet Enterprise for team use. Perplexity's enterprise page says the browser can answer queries, navigate sites, summarize pages, respond to email, send calendar invites, and adds governance features including prompt-injection protection, SOC 2 Type II and HIPAA compliance, CrowdStrike-based controls, MDM deployment, telemetry, and audit logs.
The FBI’s Seattle Division posted a victim-information page for a Steam malware investigation, saying users were primarily targeted between May 2024 and January 2026 and naming seven affected games including PirateFi and BlockBlasters.
Google said it is pairing new funding with AI-powered security tooling to help open source maintainers respond faster as AI increases both vulnerability discovery and attack pressure. The announcement combines a collective $12.5 million pledge through Alpha-Omega with wider use of tools such as Big Sleep, CodeMender, and Sec-Gemini.
OpenAI said on March 9, 2026 that it plans to acquire Promptfoo. The company said Promptfoo's technology will strengthen agentic security testing and evaluation inside OpenAI Frontier, while Promptfoo remains open source under its current license and existing customers continue to receive support.