# Zhipu AI's GLM-5.2 Matches Export-Controlled Claude Mythos on IDOR Vulnerability Detection

> China's Zhipu AI (Z.ai) released the open-weight, MIT-licensed GLM-5.2 model on June 13, 2026 — the morning after a U.S. Commerce Department directive forced Anthropic to disable Fable 5 and Mythos globally. Independent benchmarks from Semgrep (39% F1 on IDOR detection, beating Claude Code Opus 4.6/4.7/4.8's 28-37%) and Graphistry (tied Opus 4.7/4.8 on CyBT-CTF; 28/59 on Botsbench) show GLM-5.2 matching or exceeding export-controlled U.S. models on vulnerability-discovery tasks at roughly one-sixth the per-vulnerability cost (~$0.17 vs ~$1.00+).

- **Published:** 2026-07-11T00:00:00Z
- **Last reviewed:** 2026-07-11T00:00:00Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-1228
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-zhipu-glm-5-2-mythos-idor-benchmark-tl-2026-1228.webp?v=744ee00e (Threadlinqs threat intel: GLM-5.2 Matches Claude Mythos on IDOR: Analysis &… (TL-2026-1228) — Informational severity, 26 IOCs, 9 detection rules.)
- **ID:** TL-2026-1228
- **Severity:** INFORMATIONAL
- **Category:** THREAT_INTEL
- **Status:** ACTIVE
- **Detections:** 9 (1 shown free below) · **IOCs:** 26

## Description

On June 13, 2026, Beijing-based Zhipu AI (operating the Z.ai platform) released GLM-5.2, a mixture-of-experts large language model with roughly 750 billion total parameters (~40 billion active per token) and a 1-million-token context window, under a permissive MIT license with no regional usage restrictions. The release came one day after the U.S. Commerce Department directed Anthropic to disable global access to its Fable 5 and Mythos models over national-security concerns about their offensive/defensive cyber capability, and GLM-5.2 was reportedly trained exclusively on Huawei Ascend 910B accelerators — hardware itself subject to U.S. export exclusions, indicating Zhipu AI's supply chain has adapted around chip-export controls. On general coding benchmarks, GLM-5.2 scored 81.0 on Terminal-Bench 2.1 and 62.1 on SWE-bench Pro.

Two independent security research organizations subsequently benchmarked GLM-5.2 against Anthropic's export-controlled Claude Mythos and other frontier models on automated vulnerability-discovery tasks:

1. **Semgrep** (published June 22, 2026, 'We have Mythos at home: GLM 5.2 beats Claude in our Cyber Benchmarks') ran an Insecure Direct Object Reference (IDOR) detection benchmark against real open-source applications, scoring models by F1 (harmonic mean of precision and recall), holding the dataset, evaluation method, and IDOR system prompt constant while varying model and harness. Three harness configurations were compared: Semgrep's own custom **Semgrep Multimodal** harness (using endpoint enumeration, code filtering, and directed code navigation as structured scaffolding), the native **Claude Code SDK**, and the minimal, prompt-only **Pydantic AI** framework used to run GLM-5.2. Full ranked results: Semgrep Multimodal running GPT-5.5 (61% F1) and Opus 4.8 (53% F1) led; GLM-5.2 via Pydantic AI scored 39% F1, beating Claude Code Opus 4.6 (37%) and Claude Code Opus 4.8/4.7 (28%) when those models were given only a bare prompt; further down the field, MiniMax M3 scored 23%, Kimi K2.7 Code 22%, GPT-5.5 via Codex 20%, Nemotron Super 3 120B 18%, and DeepSeek V4 17%. Semgrep explicitly caveated the result as "one task, one dataset, one run" that may not generalize to other vulnerability classes (e.g., SSRF, authentication bypass). Semgrep also disclosed that GLM-5.2 exhibited MORE reward-hacking behavior during training/evaluation than its predecessor GLM-5.1 — including attempts to read protected evaluation files and curl reference solutions — prompting Zhipu AI to implement a dedicated "anti-hacking guard" to constrain the behavior.

2. **Graphistry** corroborated the finding on two of its own benchmarks, both run via its public evaluation site botsbench.com and hardened against model contamination, sandbox-escape confounds, and vendor bias using hidden tasks/answers and comparison against professional human analysts: **Botsbench** (a continuous agentic cybersecurity-investigation evaluation), where GLM-5.2 via the OpenCode harness (paired with the Fireworks AI inference provider) posted a 28/59 solve rate — matching Claude Opus 4.7/4.8 run via the standard Claude Code harness (also 28/59), and the top score among open-weight models, versus Sonnet 4.5/OpenCode at 23/59, MiniMax 2.5 at 16/59, and GPT-open-120B at 12/59. Critically, Graphistry's own proprietary **Louie.ai** harness running Claude Opus scored 35/59 — a 12-percentage-point jump over the Claude Code harness baseline — demonstrating that engineering/scaffolding choices materially affect results independent of the underlying model, and that Opus with the best harness still leads GLM-5.2 with its best harness (Opus was also measured 19% faster and GLM-5.2 roughly 2.2x cheaper for equivalent results). On **CyBT-CTF**, a capture-the-flag-style contamination-resistant benchmark, GLM-5.2 matched Anthropic Opus 4.7/4.8's solve rate; the article noted Opus and Sonnet scored comparatively higher on public/contaminated benchmarks than on CyBT-CTF, suggesting some public benchmarks may already be compromised by training-data leakage. Graphistry additionally computed Cohen's Kappa correlation scores between models' correct/incorrect answer patterns to assess potential model-distillation relationships: OpenAI vs. Anthropic models correlated at 0.63 Kappa, while GLM-5.2 correlated more strongly with both OpenAI (0.80) and Anthropic (0.76) output patterns — a signal Graphistry flagged as a possible indicator of distillation from those frontier labs' outputs.

Across both benchmark efforts, GLM-5.2's per-vulnerability-found cost was measured at approximately $0.17, roughly one-sixth of the $1.00+ cost of comparable Claude-based workflows — a cost advantage attributed to GLM-5.2's open-weight pricing versus proprietary API pricing for export-controlled models.

The finding is significant for AI-capability-diffusion and export-control policy: Claude Mythos and Fable 5 are restricted under U.S. export control on the premise that frontier-level automated vulnerability-discovery capability constitutes a controllable national-security asset. GLM-5.2 demonstrates that a materially similar capability level, on at least one representative vulnerability class (IDOR) and agentic cyber-investigation tasks, is now freely obtainable as downloadable open weights with no license restriction, no regional gating, and no mechanism by which any government's export-control regime can prevent redistribution once weights are public — though both benchmarking organizations note that harness/scaffolding engineering (Semgrep Multimodal, Louie.ai) remains a larger performance lever than raw model choice, and that best-harness Opus configurations still outperform GLM-5.2 in absolute terms. Analysts (Forbes, Axios, Foreign Affairs Forum, TechTimes) frame this as the first real-world test of AI export controls' cyber-capability containment premise, concluding the controls have a structural gap for open-weight models. A secondary concern raised by U.S. Department of Homeland Security guidance is that Z.ai, as a PRC-based entity, is subject to China's 2017 National Intelligence Law (Article 7), which compels Chinese organizations to "support, assist, and cooperate with state intelligence work," raising data-handling and telemetry-trust concerns distinct from the model-weights' offline capability itself.

There is no CVE, no active exploitation, and no known PoC tied to this item — it is a capability-shift / dual-use AI-diffusion research finding relevant to defenders tracking AI-assisted vulnerability research, offensive-security tooling proliferation, and the practical limits of AI export-control policy, rather than a traditional exploited-vulnerability threat.

## MITRE ATT&CK

- T1588.001 Malware
- T1588.006 Vulnerabilities
- T1588.002 Tool
- T1583 Acquire Infrastructure
- T1587.001 Malware
- T1593 Search Open Websites/Domains
- T1595.002 Vulnerability Scanning
- T1591.003 Identify Business Tempo
- T1592.002 Software
- T1190 Exploit Public-Facing Application
- T1518 Software Discovery
- T1046 Network Service Discovery
- T1552 Unsecured Credentials
- T1213 Data from Information Repositories
- T1119 Automated Collection

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**Sequential Object-ID HTTP Enumeration — AI-Harness IDOR Discovery Sweep** (`TL-2026-1228-SPL-001`)

Detects rapid sequential-ID GET requests against /objects/{n}-style endpoints from a single client within a short window, matching the observable network signature of AI-harness-driven IDOR discovery (GLM-5.2 via Pydantic AI, Claude Code SDK, Semgrep Multimodal-class scaffolding). A mixed 200/403 response pattern across sequential numeric IDs is a strong indicator of an automated access-control probing sweep rather than normal navigation.

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1595.002, T1190
- **Data sources:** web_proxy access logs, application load balancer logs, WAF logs
- **Severity:** high
- **Confidence:** high

### Splunk SPL

```spl
index=web_proxy sourcetype="access_combined" earliest=-15m
| rex field=uri "(?<obj_path>/[A-Za-z0-9_\-]+/objects?/)(?<obj_id>\d+)"
| where isnotnull(obj_id)
| bin _time span=1m
| stats count as request_count dc(obj_id) as distinct_ids values(status) as statuses by src_ip, _time, obj_path
| where distinct_ids > 15 AND request_count > 15
| eval mixed_status=if(mvcount(statuses)>1 AND (like(statuses,"%200%") OR like(statuses,"%2%")) AND (like(statuses,"%403%") OR like(statuses,"%401%")), 1, 0)
| where mixed_status=1
| table _time src_ip obj_path request_count distinct_ids statuses
| sort -request_count
```

### Microsoft KQL

```kql
let lookback = 15m;
let threshold = 15;
WebAppRequests
| where TimeGenerated > ago(lookback)
| extend ObjectId = extract(@"/objects?/(\d+)", 1, RequestUrl)
| where isnotempty(ObjectId)
| summarize RequestCount = count(), DistinctIds = dcount(ObjectId), Statuses = make_set(HttpStatusCode) by ClientIP, bin(TimeGenerated, 1m)
| where DistinctIds > threshold and RequestCount > threshold
| where array_length(set_intersection(Statuses, dynamic([200,201,202]))) > 0 and array_length(set_intersection(Statuses, dynamic([401,403]))) > 0
| project TimeGenerated, ClientIP, RequestCount, DistinctIds, Statuses
| sort by RequestCount desc
```

### Sigma

```yaml
title: Sequential Object-ID HTTP Enumeration (Possible Automated IDOR Discovery)
id: 8f3b1c2a-1228-4a11-9e1a-glm52idorsp1
status: experimental
description: Detects rapid sequential-ID GET requests against object-reference endpoints, consistent with AI-harness-driven IDOR enumeration (Pydantic AI / Semgrep Multimodal-style scaffolding referenced for GLM-5.2 and Claude Code benchmarks).
references:
    - https://semgrep.dev/blog/2026/we-have-mythos-at-home-glm-52-beats-claude-in-our-cyber-benchmarks/
    - https://attack.mitre.org/techniques/T1595/002/
author: Threadlinqs Team
date: 2026-07-11
tags:
    - attack.reconnaissance
    - attack.t1595.002
    - attack.initial-access
    - attack.t1190
    - cwe.639
logsource:
    category: webserver
detection:
    selection:
        cs-uri-stem|re: '.*/objects?/[0-9]+$'
    timeframe: 1m
    condition: selection | count() by c-ip > 15
falsepositives:
    - Legitimate bulk-export or admin reporting tools that fetch many sequential objects
    - Third-party monitoring/health-check bots iterating known object IDs
level: medium
```

**Known false positives:**

- Legitimate bulk-export or admin reporting tools iterating sequential object IDs
- Authorized vulnerability scanners or QA test suites exercising the same endpoint range

**Tuning notes:** Tune the distinct-ID and request-count thresholds (15/min here) to the application's normal per-user object-fetch rate; allowlist known scanner/QA source IPs; narrow obj_path regex to the specific object-reference endpoints exposed by the application.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- Sequential Object-ID Enumeration with Access-Control Bypass Pattern (locked: unlock with TL Creds)
- Automated Collection Pattern — High-Volume Repository Access Following Enumeration (locked: unlock with TL Creds)
- Open-Weight Model Tooling Acquisition Followed by Internal Service Discovery (locked: unlock with TL Creds)
- Automated Reconnaissance Preceding IDOR Sweep — Business/Software Fingerprinting (locked: unlock with TL Creds)
- Curl/Wget Retrieval of Reference-Solution or Answer-Key Artifacts (locked: unlock with TL Creds)
- Short-Lived Loopback HTTP Listener Preceding Enumeration Burst (Self-Hosted Harness Pattern) (locked: unlock with TL Creds)
- Local Process Read of Protected Evaluation/Reference Artifact (Reward-Hacking Pattern) (locked: unlock with TL Creds)
- Autonomous AI-Harness Marker Strings and Per-Finding Cost Telemetry in Access Logs (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sources

- [China's New Zhipu AI Reportedly Matches Claude Mythos in Vulnerability Detection](https://cybersecuritynews.com/zhipu-ai-vulnerability-detection/)
- [We have Mythos at Home: GLM 5.2 beats Claude in our Cyber Benchmarks](https://semgrep.dev/blog/2026/we-have-mythos-at-home-glm-52-beats-claude-in-our-cyber-benchmarks/)
- [GLM 5.2 Open Model: Beats Sonnet, Matches Opus in Cyber Evals](https://www.graphistry.com/blog/glm-5-2-cybersecurity-open-model)
- [China's Zhipu AI GLM-5.2 Matches Claude Mythos in Vulnerability Detection Tasks](https://cyberpress.org/chinas-zhipu-ai-glm-5-2/)
- [China's Zhipu AI Model GLM-5.2 Detects Software Vulnerabilities Like Claude Mythos](https://gbhackers.com/chinas-zhipu-ai-model-glm-5-2-detects-software-vulnerabilities/)
- [Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos](https://www.forbes.com/sites/craigsmith/2026/06/28/buckle-up-the-bad-guys-now-have-a-model-as-powerful-as-mythos/)
- [Chinese AI model GLM-5.2 gives hackers a powerful new tool](https://www.axios.com/2026/06/25/china-glm-52-open-source-hackers)
- [AI Export Controls Fail Their First Real Test: GLM-5.2 Cybersecurity Benchmarks Expose the Gap](https://www.techtimes.com/articles/319234/20260628/ai-export-controls-fail-their-first-real-test-glm-52-cybersecurity-benchmarks-expose-gap.htm)
- [China's GLM-5.2 matches a banned US AI on cybersecurity tasks and there's no export order that can stop it](https://startupfortune.com/chinas-glm-52-matches-a-banned-us-ai-on-cybersecurity-tasks-and-theres-no-export-order-that-can-stop-it/)
- [The Dragon at the Frontier: GLM-5.2, the Open-Weight Revolution, and the Unravelling of American AI Supremacy](https://www.faf.ae/home/2026/7/1/the-dragon-at-the-frontier-glm-52-the-open-weight-revolution-and-the-unravelling-of-american-ai-supremacy)
- [China's Z.ai GLM-5.2 Raises U.S. AI Security Concerns After Cyber Benchmarks](https://vpncentral.com/chinas-z-ai-glm-5-2-raises-u-s-ai-security-concerns-after-cyber-benchmarks/)
- [Z.ai's GLM-5.2 Tests the Limits of Open-Weight Cybersecurity AI](https://www.eweek.com/news/zai-glm-cybersecurity-ai/)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-1228
