If you are not targeting a technical AI role (ML engineering, data science), this is the priority order:

1. Practical AI tool use in your specific field — 2 to 4 weeks. The single highest-value AI skill for most professionals is genuine fluency with the AI tools relevant to their domain. Not awareness of AI in general — actual working knowledge of the specific tools your field uses. A marketer needs to know content AI and analytics tools. A lawyer needs to know legal research AI. A nurse needs to know clinical AI documentation tools. Generic AI literacy is less valuable than field-specific fluency.

2. Prompt engineering — 1 to 2 weeks of deliberate practice. The quality of what you get from AI tools is largely determined by how well you direct them. Prompt formulation is a learnable skill that improves quickly with deliberate practice. Spend two weeks intentionally experimenting with different prompt structures on real work tasks and document what works.

3. Critical output evaluation — ongoing. The ability to recognize when AI output is wrong, incomplete, biased, or confidently hallucinating is the most undervalued AI skill. It doesn't have a course — it develops through use combined with domain expertise. Any professional who uses AI tools should be actively developing their 'AI error detection' instincts.

4. One structured AI literacy course — 4 to 10 hours total. Andrew Ng's AI for Everyone (Coursera, free to audit), Google's AI Essentials, or Microsoft's AI Fundamentals provide conceptual grounding that makes the practical skills above more effective. Do this after the practical work, not before.

If you are targeting ML engineering, data science, or similar technical AI roles, the priority order is different:

1. Python programming — 6 to 12 weeks for functional proficiency. Python is the foundational language of AI/ML work. Start with Python basics, then move immediately to numpy, pandas, and matplotlib — the data manipulation and visualization libraries that form the foundation of data science work.

2. Statistics and linear algebra fundamentals — 4 to 8 weeks. You do not need a math degree, but you need functional understanding of probability, distributions, linear algebra basics, and calculus concepts. Khan Academy, 3Blue1Brown's Essence of Linear Algebra, and StatQuest with Josh Starmer are excellent free resources.

3. Machine learning fundamentals — 8 to 16 weeks. Andrew Ng's Machine Learning Specialization (Coursera) or fast.ai's Practical Deep Learning are the two most recommended starting points. Complete one end-to-end ML project during this period.

4. Portfolio projects — ongoing. After the foundations, the highest-value investment is building and publishing ML projects on GitHub. Kaggle competitions provide structured problems with public leaderboards. Two to three completed projects with good documentation is more compelling to technical hiring managers than additional coursework.

The AI skills landscape includes a lot of content that doesn't deliver proportionate career value:

Most AI vendor certifications: AWS, Google, and Microsoft all offer AI certifications. These are somewhat useful for roles specifically involving those platforms, but they are not general AI credentials. Do not spend significant time on vendor certifications until you have a specific role in mind that values them.

AI strategy courses aimed at executives: These are useful for senior leaders making AI adoption decisions, not for most professionals. The content is too high-level to be practically useful for career development.

Course stacking without project work: Ten courses and no projects is worth less than three courses and two projects. Whatever you learn, apply it. The application is what builds the credential.

Cutting-edge model papers before fundamentals: Reading about GPT architecture before understanding linear regression is backwards. Follow the priority sequence; the advanced content will make more sense when the foundations are solid.