Tricks from the Past deployed via signed ChatGPT and Claude binaries
2026-10-27T20:52:17Z•78320bbb8a2d2a0df8b52f920994544611053f64fdefd43f4d40df7d09ee79e3
CVE-2026-35029CVE-2026-59823AI-securityChatGPTClaudeLiteLLMantidetect-browsersauthorization-bypassbrowser-credentialscryptocurrency-walletsdata-breachdata-leakagedeveloper-credentialsfake-softwareinfluence-operationsinfostealermalwareonline-fraudphishingprompt-injectionsigned-binariesthreat-intelligence
What happened
Check Point Research RSS entries covering October–September 2026 cybersecurity developments, including malware campaigns abusing signed fake ChatGPT and Claude binaries, infostealers targeting AI tools and developer credentials, AI-enabled fraud and influence operations, LiteLLM authorization vulnerabilities (CVE-2026-35029 and CVE-2026-59823), prompt-injection techniques, ChatGPT sandbox data leakage, and multiple reported breaches and attacks.
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- checkpoint_research
- Record identifier
- 78320bbb8a2d2a0df8b52f920994544611053f64fdefd43f4d40df7d09ee79e3
- Enrichment time
- 2026-10-27T20:52:17Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.