Latest cybersecurity reporting from selected sources.

OpenAI's first security hire, Ari Herbert-Voss, thinks more automated bug finding will improve security without costing jobs Black Hat Asia Open source models can find bugs as effectively as Anthropic's Mythos, according to Ari Herbert-Voss, CEO of AI-powered security startup RunSybil and OpenAI's first security hire.…

Agents Fuel Digital Risk Protection, Open-Source Intel Adoption in Regulated SpacesOuttake will invest $40 million to grow its automated platform for digital risk protection and open-source threat intelligence. CEO Alex Dhillon says the New York-based startup's agent-led model stands apart by replacing manual labor with scalable AI workflows.

Update Chainlit to the latest version ASAP Two "easy-to-exploit" vulnerabilities in the popular open-source AI framework Chainlit put major enterprises' cloud environments at risk of leaking data or even full takeover, according to cyber-threat exposure startup Zafran.…

Secure-by-Design Startup Uses AI Agents to Safeguard Containers, VMs and LibrariesCloud security startup Echo has closed a $35 million Series A funding round to boost development of its AI-native OS. The platform starts with secure container images and aims to extend to VMs and libraries, helping enterprises minimize risk from open-source software.

Shanghai Firm Bets on Open-Source Strategy, Efficiency ClaimsShanghai artificial intelligence startup MiniMax released a new open-source large language model, positioning it as a direct competitor to American and other Chinese models. MiniMax says its model performs competitively on benchmark tests against leading proprietary and open models.

New Open-Source Model Rivals OpenAI, While Treading Beijing's Red LineArtificial intelligence startup DeepSeek released Thursday an updated version of its flagship reasoning model months after its Chinese origin sent shockwaves through industry. The model is a glimpse into high-performance systems being trained and deployed under norms governed by Beijing.

Acquisition Adds Expert Team, Reachability Analysis Tech to Socket's Security StackWith Coana's team and tools, Socket aims to strengthen its platform's ability to identify actionable vulnerabilities. The integration will help security teams eliminate busywork, focusing on high-impact issues using precomputed reachability data from open source codebases.

Phylum's Product Delivers Real-Time Detection of Malicious Open-Source PackagesTo combat the rise in software supply chain attacks, Veracode has acquired Denver-area startup Phylum and its advanced tools to detect malicious open-source packages. The acquisition strengthens Veracode's software composition analysis offering and enables faster, more reliable threat mitigation.

Phylum's Product Delivers Real-Time Detection of Malicious Open-Source PackagesTo combat the rise in software supply chain attacks, Veracode has acquired Denver-area startup Phylum and its advanced tools to detect malicious open-source packages. The purchase strengthens Veracode's software composition analysis offering and enables faster, more reliable threat mitigation.

Roche Replaces Sam King, Who Joined Veracode in 2006 and Became CEO in 2019Veracode tapped product leader Brian Roche as its next CEO and tasked him with helping secure the adoption of large language models and open-source software. The Boston-area application risk management vendor appointed Roche chief executive just two days after purchasing startup Longbow Security.

The four finalists in the startup competition tackled problems in firmware security, cloud infrastructure, open source software, and vulnerability remediation.

The startup's software helps organizations secure their containers in the cloud by teasing out which packages are running and which are vulnerable.

Open source + public key generation = no alerts, says infosec startup An infosec startup says it has built an Apple Airtag clone that bypasses anti-stalking protection features while running on Apple's Find My protocol.…