Top AI Ethics Jobs in Boston, MA

Evaluate AI outputs, perform audit-based reviews, identify failure modes and trends, apply and refine guidelines, communicate findings, participate in calibrations and stakeholder syncs, and help improve evaluation frameworks for human-in-the-loop AI.

Leads trust and safety for AI products serving teachers and students. Owns model guardrails, evaluations, red-teaming, monitoring, incident response, student-facing safety, district trust documentation, and regulatory alignment. Builds safety infrastructure into product development, manages incidents and postmortems, partners with Legal and Engineering, and leads a growing Trust & Safety team.

Lead enterprise AI adoption by developing rollout plans, training programs, workflow playbooks, automation use cases, tool governance, acceptable-use policies, and risk controls. Measure adoption and business outcomes, manage an AI champions network, supervise a team, and partner with Security, Legal, Compliance, IT, Procurement, Product, and Engineering. The role focuses on safe, practical AI use across healthcare operations while addressing PHI, retention, vendor data use, HIPAA, and SOC 2 requirements.

Leads enterprise knowledge and AI transformation across Customer Success, Support, Enablement, Product, Operations, and Technology. Develops knowledge architecture, taxonomy, governance, metadata, content standards, ownership, and lifecycle processes. Builds AI-ready knowledge foundations for retrieval, self-service, copilots, and agentic workflows. Evaluates and implements AI opportunities, improves content and response quality using performance data, drives Knowledge-Centered Service practices, and leads cross-functional adoption and change without direct authority.

Advises enterprise executives on AI transformation for contact center and customer service operations. Develops AI strategies, roadmaps, business cases, operating models, and value realization frameworks. Leads executive workshops and defines use cases involving virtual agents, agent copilots, analytics, automation, and agentic AI. Evaluates platforms and technologies, provides vendor-neutral recommendations, and partners with clients, vendors, and delivery teams to guide successful implementation.

Lead end-to-end red-teaming campaigns against frontier AI models by defining threat models, designing attack scenarios and taxonomies, coordinating operators, and producing executive risk reports. Establish recurring testing cadences, track exploit families, measure severity and user harm, maintain dashboards, and partner with research, product, and operations teams to validate fixes and run regression tests.

Leads AI governance and assurance engagements, designing policies, controls, operating models, risk assessments, and compliance programs. Evaluates AI security, model governance, third-party solutions, and control effectiveness against frameworks including NIST AI RMF, ISO 42001, and the EU AI Act. Collaborates with security and engineering teams to test threats such as prompt injection, model manipulation, data poisoning, and leakage. Advises executives and clients, develops methodologies and proposals, researches emerging AI risks and regulations, and supports business development.

Evaluate and improve the safety, security, and robustness of large language models. Responsibilities include designing jailbreak, prompt injection, adversarial, and tool-use abuse tests; developing regression suites and evaluation frameworks; identifying vulnerabilities; translating findings into safety recommendations; and documenting methodologies and results for technical and non-technical stakeholders.

Stress-test large language models by crafting adversarial prompts and multi-turn scenarios that expose safety, bias, hallucination, prompt injection, and guardrail vulnerabilities. Evaluate model responses using harm taxonomies and severity rubrics, document findings, refine testing methods, and collaborate with engineers, data scientists, and researchers to improve AI safety and robustness. The role involves regular exposure to disturbing harmful content.

Cinder is building a paid expert network for specialists who test AI models for safety risks. Projects may involve red teaming, jailbreak testing, image, video, audio, agentic, and MCP-connected systems. They seek experts in areas such as child safety, CBRNE, extremism, cybersecurity, fraud, self-harm, psychology, copyright, and multilingual or culturally specific model evaluation.