Cybersecurity
Jul 20, 2026
a16z Backs Neo to Reinvent Endpoint Security for AI
Andreessen Horowitz has led the seed round for Neo, a startup building endpoint security from the ground up for a world where AI agents operate with the same privileges as human users. With three SentinelOne veterans at the helm, Neo is targeting what a16z calls the third generation of endpoint security — one that existing EDR platforms were never designed to handle.
AI
Cybersecurity
Startups
Jul 18, 2026
Defenders Weaponize Prompt Injection to Stop AI Hackers
Security researchers at Tracebit have discovered that the same prompt injection technique attackers use to hijack AI systems can be flipped into a powerful defense. By planting malicious-looking prompts alongside cloud secrets, defenders can trigger an AI agent's own safety guardrails — shutting it down cold before it causes serious damage.
AI
Cybersecurity
LLMs
Jul 18, 2026
Capital One Open-Sources VulnHunter, an AI Bug Hunter
Capital One has released VulnHunter, an open-source agentic AI security tool that scans source code using an 'attacker-first' methodology and a built-in falsification engine to dramatically cut false positives. Built on Anthropic's Claude Opus 4.8, the tool represents one of the most ambitious offensive-to-defensive AI pivots by a major financial institution. For engineering and security teams, it signals a meaningful shift in how AI-powered vulnerability detection can work at scale.
AI
Cybersecurity
Open Source
Jul 16, 2026
Hacked Code Suggests Suno Scraped YouTube for AI
A breach of Suno's internal systems has exposed source code that allegedly reveals the AI music startup scraped YouTube for training data. The findings raise serious questions about copyright compliance and data sourcing practices across the generative audio space.
AI
Music
Startups
Jul 16, 2026
GPT-Red: OpenAI's In-House Model to Hack Its Own AI
OpenAI has unveiled GPT-Red, a large language model built to automatically red-team its AI systems at scale. The tool acts as an adversarial sparring partner, probing for weaknesses that human testers might miss — and it could redefine how the industry approaches AI safety evaluation.
AI
OpenAI
AI Safety
Jul 15, 2026
Oak Emerges From Stealth With $60M for AI Agent Identity
Israeli identity management startup Oak has launched out of stealth with $60 million in seed funding, targeting the growing identity chaos introduced by AI agents. Cofounded by veteran entrepreneur Shai Morag, Oak aims to fix the foundational security gaps that autonomous AI systems are rapidly exposing.
AI
Cybersecurity
Startups
Jul 14, 2026
Cybersecurity Funding Holds Firm at $10.6B in H1 2026
Despite playing second fiddle to AI megarounds, cybersecurity startups raised $10.6 billion globally in the first half of 2026 — historically strong territory. A softer Q2 masked a wave of jumbo deals, including a $600M raise by Cyera at a $12B valuation.
Cybersecurity
Venture Capital
Startups
Jul 11, 2026
Slopsquatting: AI Hallucinations Fuel Supply-Chain Risk
AI coding assistants are hallucinating fake package names — and attackers are registering them with malicious code. This emerging threat, dubbed 'slopsquatting,' is exploiting developer trust in LLMs at scale, with hallucination rates reaching as high as 82% depending on the model.
AI hallucinations
Cybersecurity
Developer Tools
Jul 11, 2026
Two Billion-Dollar Deals Headline a Big Funding Week
AI infrastructure and cybersecurity led venture activity this week, with SambaNova and Keyfactor each landing $1 billion rounds. Quantum computing, clean energy, and crypto infrastructure also attracted significant capital, rounding out a packed week for startup funding.
Venture Capital
AI
Cybersecurity
Jul 10, 2026
OpenAI Unveils GPT-5.6 and a New Model Family
OpenAI has launched GPT-5.6 as part of a broader new family of models, promising meaningful improvements across multiple domains including cybersecurity. The release marks another step in OpenAI's rapid model cadence as competition in the frontier AI space intensifies.
AI models
Cybersecurity
Enterprise AI
Jul 8, 2026
Vermilion Cliffs Ventures Closes $25M Fund II
Solo GP Ashley Smith has closed a $25 million second fund under Vermilion Cliffs Ventures, targeting early-stage startups across AI, security, and adjacent sectors. The new vehicle continues Smith's thesis-driven approach to backing technical founders at the earliest stages.
AI
Cybersecurity
Early-Stage
Jul 8, 2026
Nine AI Tools Can Be Weaponized Into Botnets via Prompts
Researchers have found that nine of the most widely used AI tools are vulnerable to a new class of prompt injection attack capable of assembling large-scale botnets. Unlike traditional targeted injections, these 'pull' attacks require no direct contact with victims — dramatically expanding the threat surface. Here's what you need to know.
AI Security
Prompt Injection
LLM
Jul 7, 2026
The 'First' AI-Run Ransomware Attack Needed a Human
An AI agent executed the technical steps of a real-world ransomware attack for the first known time — but a human still picked the target, built the infrastructure, and handed over stolen credentials. The milestone is significant, but far less autonomous than last week's headlines implied.
Cybersecurity
Ransomware
AI Agents
Jul 1, 2026
Google Warns EU Search Data Rules Could Enable Hacks
Google's senior security staff are sounding alarms over European Commission proposals that would force it to share search data and open Android to competitors. They argue the plans contain anonymization flaws that could expose millions of users' queries — and invite cybercriminals within weeks of implementation.
Android
Big Tech
Cybersecurity
Jul 1, 2026
LastPass Hit by a Breach via a Third-Party AI Vendor
LastPass customers are dealing with yet another data breach, this time stemming from a compromise at AI business intelligence firm Klue. Meanwhile, Europol and Microsoft disrupted major infostealer operations, Australia revealed nation-state hackers lurking in critical infrastructure, and John Bolton pleaded guilty to mishandling classified data.
Anthropic
ASIO
Cybersecurity
Jun 29, 2026
China's Z.ai Claims GLM-5.2 Matches Anthropic's Mythos
Zhipu AI's open-weight GLM-5.2 is drawing attention after researchers found it rivals Anthropic's Mythos in bug-finding and cybersecurity tasks. While it still trails leading US models on general benchmarks, the release signals China is closing the AI capability gap faster than expected — raising fresh concerns in Washington.
China AI
Cybersecurity
Zhipu AI
Jun 27, 2026
OpenAI's GPT-5.6 Is Ready, White House Decides Next
OpenAI has confirmed it's delaying the public release of GPT-5.6 at the request of the Trump administration. The government will pre-approve which customers can access the models first, in what OpenAI calls a temporary — but frustrating — arrangement.
OpenAI
GPT-5.6
AI Policy
Jun 27, 2026
Pentagon Probes Leak Exposing NSC, Special Ops Officers
A misconfigured website at Peter Thiel's private events group Dialog exposed personal records of over 200 national security and military personnel — including an NSC intelligence official and an active-duty Tier 1 special operations officer. The Pentagon's operations security team is now examining the breach.
Cybersecurity
Data Breach
Dialog
Jun 27, 2026
Claude's Great Wall Problem: China's AI Underground
Despite aggressive geolocation restrictions and account bans, Chinese developers, researchers, and startups have built a thriving underground economy to access Anthropic's Claude. From Taobao account resellers to API 'transfer stations,' the cat-and-mouse game shows no signs of slowing down.
Anthropic
Claude
China
Jun 27, 2026
Trump Admin Eases Restrictions on Claude Mythos 5
The US government has eased export controls on Anthropic's most powerful AI model, Claude Mythos 5, allowing over 100 US organizations to regain access. The move follows weeks of tense negotiations between Anthropic and the Trump administration, though broader restrictions — including those on the consumer-facing Claude Fable 5 — remain in place.
Anthropic
AI Policy
Claude