Threat Intelligence

Threat Intelligencehttps://threat-intelligence.anth.us/2026-07-04T00:00:00ZAnthus AI SolutionsPractical advice for staying secure as the threat landscape shifts.Govern OpenAI Workspaces Like Access Boundarieshttps://threat-intelligence.anth.us/ai/treat-openai-accounts-like-production-infrastructure/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

OpenAI security is not prompt hygiene. Workspaces, projects, keys, connectors, and files decide who can use AI systems and what business data they can reach.

Treat Mods and Launchers Like Untrusted Codehttps://threat-intelligence.anth.us/gaming/treat-mods-and-launchers-like-untrusted-code/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Mod loaders, overlays, trainers, and private-server patches are supply chain decisions. Install them only where you are willing to rebuild.

The New Sensitive Data Estatehttps://threat-intelligence.anth.us/mission/the-new-sensitive-data-estate/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

AI/ML work concentrates value in artifacts that do not look like databases: embeddings, prompt logs, notebooks, evaluation sets, traces.

The Knowledge Base Beneath the Newsroomhttps://threat-intelligence.anth.us/inside-the-newsroom/the-knowledge-base-beneath-the-newsroom/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

A useful AI newsroom needs memory: curated sources, extracted entities, and graph structure that improves with every cycle.

The Balance of Power Is Shiftinghttps://threat-intelligence.anth.us/mission/the-balance-of-power-is-shifting/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

AI is making attack capacity cheap — more targets, more attempts, more persistence. Controls that were tolerated because capable adversaries were rare need a new threat model.

Scope OpenAI Keys to One Jobhttps://threat-intelligence.anth.us/ai/shrink-openai-api-key-blast-radius/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

An OpenAI key should represent one workload, one owner, and one clean revocation path.

Separate Game Accounts From Real Lifehttps://threat-intelligence.anth.us/gaming/separate-game-accounts-from-real-life/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Store accounts, Discord identities, recovery email, and payment methods can become paths from a compromised game account back into real life.

Make Azure Admin Access Expirehttps://threat-intelligence.anth.us/azure/make-azure-privilege-temporary/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Standing admin access is a durable target. PIM, Conditional Access, and access reviews make privilege temporary, justified, and visible.

How to Play Games Securelyhttps://threat-intelligence.anth.us/gaming/how-to-play-games-securely/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Games, mods, launchers, and anti-cheat drivers are software from strangers. Give them their own blast radius, away from banking, work, and recovery accounts.

How Our Newsroom Learnshttps://threat-intelligence.anth.us/inside-the-newsroom/how-our-newsroom-learns/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Attackers no longer get tired. Defenders need the same persistence in how they learn — expert judgment steering, AI doing the patient watching.

From Signals to Practical Advicehttps://threat-intelligence.anth.us/inside-the-newsroom/from-signals-to-practical-advice/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Research, reporting, drafting, and review are different jobs with different constraints. The handoffs are what keep evidence and uncertainty intact.

From Lessons Learned to Defenses Checkedhttps://threat-intelligence.anth.us/mission/from-lessons-learned-to-defenses-checked/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

The point of threat intelligence is not to admire the attack. Every lesson someone else paid for should become a check you can run.

Classify Azure Data Where Access Paths Reach Ithttps://threat-intelligence.anth.us/azure/find-sensitive-data-paths-in-azure/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

A catalog tells you where sensitive data may live. Security work starts when classification meets identity and network reality.

Find PII Risk in Your S3 Bucketshttps://threat-intelligence.anth.us/aws/find-pii-risk-in-s3-buckets/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Which S3 buckets hold PII, and which of those can actually be reached? The overlap is where the work is.

Control ChatGPT Connectors and Actionshttps://threat-intelligence.anth.us/ai/control-chatgpt-workspace-access-and-connected-data/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Once ChatGPT can reach drives, repositories, and actions, workspace security is about what the workspace can reach and do — not just who can log in.

Build the AWS Findings Pipelinehttps://threat-intelligence.anth.us/aws/build-the-aws-exposure-control-stack/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

Detection is the easy half. A findings pipeline routes signals into ownership, escalation, and closure — so exposure actually goes down.

Map Azure Attack Paths Through Identityhttps://threat-intelligence.anth.us/azure/audit-azure-blast-radius-before-attackers-do/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

In Azure the center of gravity is identity. The question is which identities and workloads can cross boundaries into sensitive systems.

Turn AWS Findings Into an Exposure Queuehttps://threat-intelligence.anth.us/aws/audit-aws-exposure-before-attackers-do/2026-07-04T00:00:00Z2026-07-04T00:00:00Z

AWS risk rarely arrives as one finding. It shows up when identity, data, key policy, and logging gaps line up across accounts — and the work is to see the line-up first.