CTIPilot
AI-generated · no human review · verify critical claims against the linked source. how it works →

GTIG AI Threat Tracker (May 2026)

report · report:gtig-ai-threat-tracker-may-2026

Google Threat Intelligence Group AI Threat Tracker of 2026-05-11: a criminal zero-day 2FA-bypass exploit GTIG believes was AI-developed, prepared for planned mass exploitation that disclosure may have prevented; LLM-generated decoy code in CANFAIL and LONGSTREAM; the Gemini-driven Android backdoor PROMPTSPY; PRC and DPRK actors using Gemini for vulnerability research; LLM account pooling; and TeamPCP supply-chain attacks on AI components.

Coverage
1
first 2026-05-12 → last 2026-10-21
Latest activity
2026-05-12
GTIG finds a criminal zero-day exploit it believes was built with AI; its discovery may have prevented mass…
Peak priority
notable
1 notable
Targets
public-sector
sectors: public-sector, defense, technology
Sources cited
5
5 hosts

Defender insights

What each entry about GTIG AI Threat Tracker (May 2026) tells a defender to do, newest first.

2026-05-12NOTABLEGTIG finds a criminal zero-day exploit it believes was built with AI; its discovery may have prevented mass exploitation

Exposure · detection

Story timeline

  1. 2026-05-12GTIG AI Threat Tracker (May 2026): an AI-developed zero-day 2FA bypass prepared for mass exploitation, AI decoy code in Russia-nexus malware and PROMPTSPY's Gemini agent
    deep-diveGTIG finds a criminal zero-day exploit it believes was built with AI; its discovery may have prevented mass exploitation
ATT&CK techniques (10 across 8 tactics)

10 techniques observed across 1 entry about this entity, derived from entry metadata and body evidence, never asserted without a published entry behind it · pinned to MITRE ATT&CK v20.1 · compare on the matrix · Navigator layer (JSON)

  • ReconnaissanceGather Victim Org Information · Gather Victim Host Information: Hardware
  • Resource DevelopmentDevelop Capabilities: Exploits · Obtain Capabilities: Artificial Intelligence
  • Initial AccessValid Accounts · Exploit Public-Facing Application · Supply Chain Compromise: Compromise Software Supply Chain
  • PersistenceValid Accounts
  • Privilege EscalationValid Accounts
  • StealthObfuscated Files or Information: Junk Code Insertion · Valid Accounts
  • Credential AccessUnsecured Credentials
  • Command and ControlProxy: Multi-hop Proxy

Reconnaissance TA0043

T1591Gather Victim Org Information×1

Adversaries may gather information about the victim's organization that can be used during targeting. Information about an organization may include a variety of details, including the names of divisions/departments, specifics of business operations, as well as the roles and responsibilities of key employees.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

T1592.001Gather Victim Host Information: Hardware×1

Adversaries may gather information about the victim's host hardware that can be used during targeting. Information about hardware infrastructure may include a variety of details such as types and versions on specific hosts, as well as the presence of additional components that might be indicative of added defensive protections (ex: card/biometric readers, dedicated encryption hardware, etc.).

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Resource Development TA0042

T1587.004Develop Capabilities: Exploits×1

Adversaries may develop exploits that can be used during targeting. An exploit takes advantage of a bug or vulnerability in order to cause unintended or unanticipated behavior to occur on computer hardware or software. Rather than finding/modifying exploits from online or purchasing them from exploit vendors, an adversary may develop their own exploits. Adversaries may use information acquired via Vulnerabilities to focus exploit development efforts. As part of the exploit development process, adversaries may uncover exploitable vulnerabilities through methods such as fuzzing and patch analysis.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

T1588.007Obtain Capabilities: Artificial Intelligence×1

Adversaries may obtain access to generative artificial intelligence (AI) tools, such as large language models (LLMs), to aid various techniques during targeting. These tools may be used to inform, bolster, and enable a variety of malicious tasks, including creating scripts, assisting social engineering, developing payloads, and to Query Public AI Services in support of Reconnaissance.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Initial Access TA0001

T1078Valid Accounts×1

Adversaries may obtain and abuse credentials of existing accounts as a means of gaining Initial Access, Persistence, Privilege Escalation, or Defense Evasion. Compromised credentials may be used to bypass access controls placed on various resources on systems within the network and may even be used for persistent access to remote systems and externally available services, such as VPNs, Outlook Web Access, network devices, and remote desktop. Compromised credentials may also grant an adversary increased privilege to specific systems or access to restricted areas of the network. Adversaries may choose not to use malware or tools in conjunction with the legitimate access those credentials provide to make it harder to detect their presence.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

T1190Exploit Public-Facing Application×1

Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network. The weakness in the system can be a software bug, a temporary glitch, or a misconfiguration.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

T1195.002Supply Chain Compromise: Compromise Software Supply Chain×1

Adversaries may manipulate application software prior to receipt by a final consumer for the purpose of data or system compromise. Supply chain compromise of software can take place in a number of ways, including manipulation of the application source code, manipulation of the update/distribution mechanism for that software, or replacing compiled releases with a modified version.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Persistence TA0003

T1078Valid Accounts×1

Adversaries may obtain and abuse credentials of existing accounts as a means of gaining Initial Access, Persistence, Privilege Escalation, or Defense Evasion. Compromised credentials may be used to bypass access controls placed on various resources on systems within the network and may even be used for persistent access to remote systems and externally available services, such as VPNs, Outlook Web Access, network devices, and remote desktop. Compromised credentials may also grant an adversary increased privilege to specific systems or access to restricted areas of the network. Adversaries may choose not to use malware or tools in conjunction with the legitimate access those credentials provide to make it harder to detect their presence.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Privilege Escalation TA0004

T1078Valid Accounts×1

Adversaries may obtain and abuse credentials of existing accounts as a means of gaining Initial Access, Persistence, Privilege Escalation, or Defense Evasion. Compromised credentials may be used to bypass access controls placed on various resources on systems within the network and may even be used for persistent access to remote systems and externally available services, such as VPNs, Outlook Web Access, network devices, and remote desktop. Compromised credentials may also grant an adversary increased privilege to specific systems or access to restricted areas of the network. Adversaries may choose not to use malware or tools in conjunction with the legitimate access those credentials provide to make it harder to detect their presence.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Stealth TA0005

T1027.016Obfuscated Files or Information: Junk Code Insertion×1

Adversaries may use junk code / dead code to obfuscate a malware’s functionality. Junk code is code that either does not execute, or if it does execute, does not change the functionality of the code. Junk code makes analysis more difficult and time-consuming, as the analyst steps through non-functional code instead of analyzing the main code. It also may hinder detections that rely on static code analysis due to the use of benign functionality, especially when combined with Compression or Software Packing.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

T1078Valid Accounts×1

Adversaries may obtain and abuse credentials of existing accounts as a means of gaining Initial Access, Persistence, Privilege Escalation, or Defense Evasion. Compromised credentials may be used to bypass access controls placed on various resources on systems within the network and may even be used for persistent access to remote systems and externally available services, such as VPNs, Outlook Web Access, network devices, and remote desktop. Compromised credentials may also grant an adversary increased privilege to specific systems or access to restricted areas of the network. Adversaries may choose not to use malware or tools in conjunction with the legitimate access those credentials provide to make it harder to detect their presence.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Credential Access TA0006

T1552Unsecured Credentials×1

Adversaries may search compromised systems to find and obtain insecurely stored credentials. These credentials can be stored and/or misplaced in many locations on a system, including plaintext files (e.g. Shell History), operating system or application-specific repositories (e.g. Credentials in Registry), or other specialized files/artifacts (e.g. Private Keys).

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Command and Control TA0011

T1090.003Proxy: Multi-hop Proxy×1

Adversaries may chain together multiple proxies to disguise the source of malicious traffic. Typically, a defender will be able to identify the last proxy traffic traversed before it enters their network; the defender may or may not be able to identify any previous proxies before the last-hop proxy. This technique makes identifying the original source of the malicious traffic even more difficult by requiring the defender to trace malicious traffic through several proxies to identify its source.

Evidence: 2026-05-12/gtig-ai-threat-tracker-may-2026-first-confirmed-ai-generated · ATT&CK page ↗

Entries about GTIG AI Threat Tracker (May 2026) (1)

2026-05-12 · view entry permalink →

NOTABLEupdatedNATOB2

GTIG AI Threat Tracker (May 2026): an AI-developed zero-day 2FA bypass prepared for mass exploitation, AI decoy code in Russia-nexus malware and PROMPTSPY's Gemini agent

Google Threat Intelligence Group's AI Threat Tracker of 2026-05-11, drawn from Mandiant incident response engagements, Gemini and GTIG's own research, follows its February 2026 report on AI-related threat activity (GTIG, 2026-05-11).

An AI-developed zero-day prepared for mass exploitation

GTIG observed prominent cybercrime actors partnering to plan a mass vulnerability-exploitation operation around a zero-day exploit, implemented in a Python script, that bypasses two-factor authentication on a popular open-source, web-based system administration tool; GTIG worked with the vendor to disclose the flaw and disrupt the activity, and says its counter-discovery may have prevented the exploit's use (GTIG, 2026-05-11). The bypass requires valid user credentials and stems from a high-level semantic logic flaw, a trust assumption the developer hardcoded, rather than from memory corruption or input sanitisation (GTIG, 2026-05-11). GTIG does not believe Gemini was used but has high confidence that an AI model supported the discovery and weaponisation, citing abundant educational docstrings, a hallucinated CVSS score and a textbook Pythonic structure with detailed help menus and a clean _C ANSI colour class (GTIG, 2026-05-11). Google did not name the tool (The Hacker News, 2026-05-11). Contradiction: The Hacker News calls this the first time AI was put to use in the wild for exploit generation (The Hacker News, 2026-05-11), whereas GTIG says the planned mass exploitation may have been prevented (GTIG, 2026-05-11).

State actors using AI for vulnerability research and infrastructure

UNC2814 prompted Gemini with expert security personas to support vulnerability research into embedded targets including TP-Link firmware and Odette File Transfer Protocol (OFTP) implementations, and APT45 sent thousands of repetitive prompts that recursively analyse different CVEs and validate proof-of-concept exploits (GTIG, 2026-05-11). APT27 used Gemini to develop a fleet-management application likely meant for an operational relay box network, with the hop count hardcoded to three and mobile Wi-Fi and router devices among the supported types (GTIG, 2026-05-11). Actors also prompt models for the organisational hierarchies and third-party relationships of large enterprises to tailor phishing lures, one tried to identify the exact computer model a high-value target used, and a suspected PRC-nexus actor ran the agentic tools Hexstrike and Strix against a Japanese technology firm and an East Asian cybersecurity platform (GTIG, 2026-05-11).

Malware with AI-generated decoys and an AI-driven Android backdoor

Russia-nexus activity against Ukrainian organisations delivers CANFAIL and LONGSTREAM, which use LLM-generated decoy code to obfuscate their malicious functionality: CANFAIL's source carries LLM comments labelling unused filler blocks, and LONGSTREAM queries the system's daylight-saving status 32 times to pad the script with benign-looking activity (GTIG, 2026-05-11). PROMPTFLUX and HONESTCUE, which call the Gemini API to generate or obfuscate code, featured in GTIG's earlier reports (GTIG, 2026-05-11). The Android backdoor PROMPTSPY, first identified by ESET, runs an agent module that serialises the visible user interface through the Accessibility API, sends it to the gemini-2.5-flash-lite model in an HTTP POST and turns the JSON reply into simulated taps and swipes; it can replay captured PINs or lock patterns, lays an invisible overlay over the uninstall button and can rotate its Gemini API keys and VNC relay through its C2 (GTIG, 2026-05-11).

Pooled LLM access and attacks on AI components

UNC6201 tried a public script that registers and immediately cancels premium LLM accounts, and UNC5673, a PRC-nexus cluster with notable overlaps with TEMP.Hex that has targeted government sectors mainly in South and Southeast Asia, uses Claude-Relay-Service and CLI-Proxy-API to pool Gemini, Claude and OpenAI accounts (GTIG, 2026-05-11). TeamPCP (UNC6780) gained initial access through compromised PyPI packages and malicious pull requests to GitHub repositories including those of Trivy, Checkmarx, LiteLLM and BerriAI, embedded the SANDCLOCK credential stealer to extract AWS keys and GitHub tokens from build environments and monetised them with ransomware and data-theft extortion groups; GTIG adds that the LiteLLM compromise could expose considerable AI API secrets (GTIG, 2026-05-11).

For the first time, GTIG has identified a threat actor using a zero-day exploit that we believe was developed with AI. The criminal threat actor planned to use it in a mass exploitation event but our proactive counter discovery may have prevented its use.

The vulnerability can be classified as a 2FA bypass, though it requires valid user credentials in the first place.

Analysis confirms the use of CANFAIL and LONGSTREAM, which utilize LLM-generated decoy code to obfuscate their malicious functionality.

For example, they employ "Claude-Relay-Service" to aggregate multiple Gemini, Claude, and OpenAI accounts, enabling account pooling and cost-sharing.

TeamPCP gained initial access through compromised PyPI packages and malicious pull requests to these GitHub repositories.

Google Threat Intelligence Group 2026-05-11

marking the first time the technology has been put to use in the wild in a malicious context for vulnerability discovery and exploit generation.

The Hacker News 2026-05-11
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationentitiesevidenceheadlineprioritysourcessourcing_notesummarytechniquestitle

GTIG reports the first zero-day exploit it believes was developed with AI, prepared by criminals for a planned mass-exploitation event that its counter-discovery may have prevented, not an exploit confirmed in use in the wild; the 2FA bypass also requires valid user credentials (GTIG, 2026-05-11). CANFAIL and LONGSTREAM do not call an LLM at runtime: they carry LLM-generated decoy code to obfuscate their functionality, and the report's example of malware driven by an LLM at runtime is the Android backdoor PROMPTSPY (GTIG, 2026-05-11). UNC5673 overlaps with TEMP.Hex rather than being it, and uses relay tooling to pool LLM accounts; the report does not describe stolen LLM API keys as a primary objective or a criminal market for them, and its only point on AI API secrets is that the LiteLLM compromise could expose them (GTIG, 2026-05-11). TeamPCP entered through compromised PyPI packages and malicious pull requests to GitHub repositories, and embedded SANDCLOCK to take AWS keys and GitHub tokens from build environments (GTIG, 2026-05-11).

threat12 May 05:00Zmulti-sourceOpen finding →

Co-occurring entities

Derived: referenced by the same focused operational entries (weekly summaries and report roundups don't count); ×N counts the shared entries.

Where this entity is cited

  • Deep dive1

Source distribution

  • cloud.google.com1 (20%)
  • helpnetsecurity.com1 (20%)
  • securityweek.com1 (20%)
  • thehackernews.com1 (20%)
  • theregister.com1 (20%)