Certifications · 13 listed · 8 issuers
Which AI certification is right for you?
What each certification covers, who it is for, and the key terms to know before you sit the exam.
Exams change often. Details here were checked against each issuer's own pages (the date is on every certification), but always confirm on the issuer's site before you book. We are not affiliated with any issuer, and prices are left out because they vary.
13 certifications
Shows you understand AI, machine learning and generative AI concepts, and when to use AWS's AI services. You use AI solutions rather than build them.
Shows you can build production-grade generative AI applications on AWS: connecting models and data, keeping them safe, and testing and tuning them.
Business-level certification: what generative AI is, what Google Cloud offers, how to improve model output and how to adopt AI well. No hands-on coding.
Shows you can build, deploy and run machine learning solutions on AWS. The updated version adds generative AI and agent topics.
Anthropic's role-based credentials for people who work with Claude: Associate, Developer, Architect and Architect Professional. The developer and architect paths cover the API, tool use and agent systems.
Shows you can design and build LLM-powered solutions on Databricks, from preparing data to deploying and monitoring.
Shows you can design, build, ship and monitor machine learning and AI solutions on Google Cloud, including MLOps and responsible AI.
For people who guide how organisations use AI responsibly: the laws that apply, ethical impact, risk management and governance frameworks.
Shows you can design, build and deploy AI solutions and agents on Azure with Python and Microsoft Foundry.
Entry-level Microsoft certification. The current exam covers AI concepts and building simple AI solutions with Microsoft Foundry.
Entry-level credential for the basics of building, connecting and maintaining apps that use large language models, with NVIDIA tools.
Shows you can design, build, deploy and govern advanced agent systems, including multi-agent teamwork, scaling and safeguards.
A course and certification where you practise real tasks inside ChatGPT and earn a credential for job-ready AI skills. It can lead on to a broader OpenAI Certification.
Terms that show up again and again
If you learn these first, you will recognise a large part of the material on most of these certifications.
- Responsible AI10 certs
- Building and using AI fairly, safely, transparently and with accountability.
- Prompt engineering9 certs
- Writing instructions so a model gives better answers.
- Retrieval-augmented generation (RAG)8 certs
- Looking up relevant documents first, then giving them to the model to answer from.
- Guardrails5 certs
- Rules and filters that block unsafe inputs or outputs.
- Hallucination5 certs
- When a model states something false in a confident way.
- Large language model (LLM)5 certs
- A foundation model that reads and writes text.
- AI agent4 certs
- Software that uses a model to plan steps and use tools to reach a goal.
- AI governance4 certs
- The policies, roles and records that control how AI is built and used in an organisation.
- Bias4 certs
- Unfair skew in a model's results, often caused by skewed training data.
- Evaluation (evals)4 certs
- Tests that measure how good a model or an AI app is.