GenAI Safety & Adversarial Testing Specialist

TikTok; San Jose (CA); On-site; USD 122,000 - 272,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

401(k) with company match

Parental leave

Disability insurance

Life insurance

Wellbeing benefits

Paid holidays (10 per year)

Paid sick days (10 per year)

Paid Personal Time (17 days)

Job summary

TikTok seeks a seasoned Trust & Safety GenAI adversarial tester to uncover emerging risks in our AI models and policies. You will work across product and policy teams to identify failure modes and provide actionable mitigations, before and after launch.

A strong background in AI safety, jailbreak analysis, and cross-functional collaboration is essential. You will investigate prompts, model behavior, and abuse patterns, communicating findings clearly to engineers and policy leads.

Qualifications

• Minimum 3 years of experience in Trust & Safety, cybersecurity, risk/adversarial testing, or related fields.
• Experience with prompt testing, jailbreak analysis, LLM evaluation, or adversarial QA.
• Familiarity with AI safety risks (jailbreaks, hallucinations, bias, misuse patterns).
• Strong interest in GenAI safety and how AI systems can be compromised under adversarial conditions.
• Demonstrated ability to independently investigate ambiguous problems and produce clear conclusions.
• Ability to manage multiple priorities and collaborate across teams.

Responsibilities

• Conduct structured adversarial testing on AI models, features, and policies to identify risks.
• Explore product behavior across contexts and user journeys to find unseen failure modes.
• Investigate jailbreaks, evasions, prompt-based attacks and other adversarial techniques.
• Document findings with risk descriptions, reproduction steps, severity, and mitigations.
• Partner with policy, product, and business teams to validate mitigations and root causes.
• Support development of testing playbooks, taxonomies, and knowledge bases.
• Stay updated on emerging adversarial trends and external risk landscape.

Skills

Trust & Safety experience

Cybersecurity

Adversarial testing

Prompt testing

LLM evaluation

AI safety risks

Ambiguity investigation

Cross-functional collaboration

Job description

TikTok seeks a seasoned Trust & Safety GenAI adversarial tester to uncover emerging risks in our AI models and policies. You will work across product and policy teams to identify failure modes and provide actionable mitigations, before and after launch.

A strong background in AI safety, jailbreak analysis, and cross-functional collaboration is essential. You will investigate prompts, model behavior, and abuse patterns, communicating findings clearly to engineers and policy leads.