Fetching from the wire…
01
02
03
04
05
06
07
08
Policy2026-07-28 · source-backed
VERITAS, led by Anita Nikolich at Illinois, won a three-year $896,000 NSF Cybersecurity Innovation for Cyberinfrastructure grant. Its most concrete output is a staff role piloted at NCSA: an engineer who reviews projects pre-deployment, scans for malicious behavior, checks software vulnerabilities and explicitly evaluates how much autonomy an agent has been granted. The project also standardizes model cards and dataset datasheets and builds training that teaches students to spot poisoned data and assess agent permissions. Red-teaming practice arriving in scientific research computing, with a title attached. (Help Net Security)
Each link below shares sources, entities, or timing with this story.
Announced August 5, funding up to 10 hubs, one award per state or region, pooling advanced computing, data and software for AI-enabled research and education, with NVIDIA, AMD, Intel, Dell, Hangar and the Secunda Innovation Fund all committing support. It also funds consortium...
The MCP ecosystem now has 30 documented CVEs across six weeks spanning three attack layers: server-side injection (43%), protocol library flaws, and developer tooling vulnerabilities. Simultaneously, the AIUC-1 Consortium reports 80% of enterprises have observed risky agent be...
The AIUC-1 Consortium (with Stanford Trustworthy AI Research Lab and 40+ security executives) reports the average enterprise runs ~1,200 unofficial AI apps. Shadow AI breaches cost $670K more than standard incidents. 63% of employees paste sensitive data into personal chatbot...
The FTC's AI policy statement — mandated within 90 days of Trump's December 2025 Executive Order — is due today. Leaked drafts indicate it will apply existing consumer protection statutes (Section 5, COPPA, FCRA, ECOA) to AI systems covering AI-generated ads, training data con...
A joint study across OpenAI, Anthropic, and Google DeepMind found no single filter or classifier holds up against an adaptive attacker. (Help Net Security) Treat injection as a containment problem, not a detection one: separate trusted from untrusted text, validate output stru...
1.5B illicit AI discussions on criminal forums. 3.3B stolen credentials. Criminals building autonomous intrusion cycles. But "stitching together tools not designed as a single automated process" is still hard — the gap mirrors legitimate enterprise adoption challenges. Help Ne...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.