Red Teaming Stories

Global Stories

Red Teaming stories

AI security training surges as prompt injection tops list

AI-related training is shifting as prompt injection, model exploitation and agent hijacking shape how security teams prepare for live attacks.

By Joseph Gabriel Lagonsin • 4 min read • Yesterday

Cloud Security

Anthropic model can chain bugs into exploits, Cloudflare

The findings suggest AI-assisted bug hunting is edging closer to practical exploitation, raising the stakes for software teams racing to patch flaws.

By Mark Tarre • 4 min read • Yesterday

Phishing

IRONSCALES names Steve McKenzie Chief Operating Officer

Pressure is mounting on email security vendors as AI-driven phishing grows, and IRONSCALES has brought in a former Mimecast executive to help scale up.

By Sofiah Nichole Salivio • 4 min read • Yesterday

Digital Transformation

Synack report says vulnerability testing gap widens

Enterprises are testing only about 32% of their attack surface, leaving many assets outside regular security checks as threats grow faster.

By Mark Tarre • 4 min read • Last week

Supply Chain

Outer Hebrides cyber firm wins Scotland tech award

The award puts a remote island cyber specialist in the national spotlight as firms seek more help against rising attacks.

By Mark Tarre • 3 min read • Last week

Digital Transformation

HackerOne links validated flaws to Wiz cloud platform

Security teams may cut backlogs as validated HackerOne flaws are mapped into Wiz, linking exploit evidence to cloud assets for faster prioritisation.

By Sofiah Nichole Salivio • 4 min read • Last week

Digital Transformation

HackerOne & Wiz link validated findings to cloud risk

Security teams can now rank cloud flaws by exploitability and impact, as validated HackerOne reports feed directly into Wiz's risk graph.

By Sofiah Nichole Salivio • 3 min read • Last week

Cloud Security

Sweet launches AI red-team agent to test attack paths

Security teams under pressure to prove real exploitability can now test live production systems for attack paths rather than theoretical flaws.

By Sofiah Nichole Salivio • 4 min read • Last week

Phishing

IRONSCALES gets Anthropic verification for email defence

Verified access to Anthropic's restricted AI tools could help IRONSCALES test email defences against more realistic phishing and impersonation attacks.

By Sean Mitchell • 4 min read • Last week

Data Protection

Integrated Quantum launches Kaggle test of VEIL anonymity

The public test could bolster or undermine claims that VEIL can anonymise sensitive AI data without letting outsiders recover the original records.

By Sean Mitchell • 4 min read • Last week

Digital Transformation

CyberCX report finds 29% of tests exposed severe flaws

AI systems and social engineering tests proved especially risky, as CyberCX found severe weaknesses in half and 77% of cases respectively.

By Sean Mitchell • 4 min read • Last week

Malware

AI now routine in cyber attacks, Google report finds

Security teams face a broader threat as criminals and state-backed actors use generative AI to speed hacks, phishing and malware.

By Joseph Gabriel Lagonsin • 4 min read • Last week

Data Protection

CurricuLLM launches Student Safety Centre for schools

CurricuLLM rolls out a school AI monitoring tool in Australia and New Zealand, flagging 21 harm types from academic offloading to personal revelations.

By Mark Tarre • 4 min read • This month