# Wiz tests Gemini 4 Argon on cyber threats, finds 2x more vulnerabilities

Google · 2026-10-21 · Independent evaluation

> Google cites Wiz researchers' real-world testing in which Gemini 4 Argon found more than twice as many vulnerabilities as baseline models, including a critical risk previous frontier models missed.

This update bears on 7 kinds of work in 4 markets. Highest level accepted: L2 Partial automation.

## The work it bears on

- [Assess practical performance and occupational competence](https://surviagi.com/work/education-vocational-training-competence-assessment) (Vocational skills and practical training): L2 Partial automation · Independently confirmed
- [Build research prototypes and run comparative experiments](https://surviagi.com/work/science-research-computing-engineering-engineering-experiments) (Computing and engineering research): L2 Partial automation · Independently confirmed
- [Monitor security alerts and investigate anomalies](https://surviagi.com/work/software-cybersecurity-security-detection) (Cybersecurity and privacy engineering): L2 Partial automation · Independently confirmed
- [Review code configuration and access permissions](https://surviagi.com/work/software-cybersecurity-security-review) (Cybersecurity and privacy engineering): L2 Partial automation · Independently confirmed
- [Reproduce diagnose and verify defect fixes](https://surviagi.com/work/software-software-testing-defect-diagnosis) (Software testing and defect analysis): L2 Partial automation · Independently confirmed
- [Run automated and exploratory tests](https://surviagi.com/work/software-software-testing-test-execution) (Software testing and defect analysis): L2 Partial automation · Independently confirmed
- [Design test scenarios and test data](https://surviagi.com/work/software-software-testing-test-design) (Software testing and defect analysis): L1 Assisted · Independently confirmed

## Published at

- [Source 1](https://x.com/Google/status/2112914740058181739)

---

Page: https://surviagi.com/updates/ev-bd19f1257ada78d86e854d448f40230bad51f509

Data generated 2026-10-22T00:18:52.702527+00:00; method activity-level-1; ontology 2.4.0.
Machine-proposed: no person has reviewed these readings, and the method is not calibrated.
Cite the page address together with the generation time.