The DTC beauty category is crowded. To break through, Jennifer Anniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

Let me be straight with you about something.

Most articles about AI skills in 2026 are written for people who want to become AI engineers. They tell you to learn Python, TensorFlow, neural networks, and deep learning. And then you close the tab feeling overwhelmed and behind.

But here is the truth. The vast majority of people reading this are not trying to build AI models. They are trying to keep their jobs, grow in their careers, and not get left behind while everything changes around them.

This newsletter is for those people.

Here are the only AI skills that actually matter for staying employable in 2026, whether you work in marketing, finance, operations, design, HR, or anywhere else.

Before we get into the skills, one number worth sitting with.

Candidates with AI-related skills command, on average, an advertised salary 23% higher than otherwise comparable candidates without those skills. That is higher than the premium for a Master's degree.

Read that again. AI skills now outperform formal degrees in what employers are willing to pay. And the gap is only widening.

Over the last three years, mentions of AI in job listings have increased by over 600% in the US. Yet true expertise remains in short supply.

That gap between what employers need and what candidates can offer is your opportunity. You do not need to be an expert. You just need to be ahead of most people. And right now, that bar is lower than you think.

This is the single most important skill for anyone who uses AI tools in their work, which by 2026 means almost everyone.

Prompt engineering simply means knowing how to talk to AI so it gives you useful outputs instead of generic ones. It is the difference between asking ChatGPT "write me an email" and getting something bland, versus giving it context, tone, audience, and goal and getting something you can actually send.

Prompt engineering improves output quality across any role. It is now a critical skill for non-technical roles, and strong prompting improves accuracy, efficiency, and output quality across every task.

The good news is this takes days to learn, not months. Start by practicing with any free AI tool. Learn to give context, set the format, specify the audience, and ask for revisions. That is genuinely most of what you need.

Skills explicitly tied to applying AI rose 109% year over year. Companies are embedding AI into daily workflows, and they are paying for human talent to make it work.

What this means practically is that employers want people who can connect AI tools to the work that actually needs doing. This does not mean coding. It means knowing how to use tools like Zapier, Make, Power Automate, or even simple AI assistants to automate repetitive tasks.

Think about the parts of your job that feel copy-paste. Drafting reports. Sorting emails. Summarising meeting notes. Formatting data. All of these can now be partially or fully automated with the right tools, and the person who figures that out first on their team becomes indispensable.

You do not need to be a data analyst. But you do need to be comfortable reading and questioning data outputs that AI tools generate.

Data fluency makes you a sharper, more independent contributor, and data skills consistently rank among the top competencies employers demand globally, with demand growing across sectors beyond tech including finance, healthcare, and marketing.

In practical terms this means being able to look at an AI-generated report or chart and ask the right questions. Where did this data come from? Does this trend make sense? What is missing from this picture? That critical thinking layer is something AI cannot do for itself, and it is what keeps humans valuable.

Employers increasingly prioritise practical AI abilities over deep technical knowledge. They want flexible, multi-skilled individuals who understand how to use AI to improve productivity across different areas.

This skill is simply staying current. Knowing that Perplexity exists for research. That Midjourney or Ideogram can produce visuals. That Notion AI can summarise documents. That Claude or ChatGPT can help draft, analyse, and brainstorm. That tools like Otter or Fireflies can transcribe meetings automatically.

You do not need to master all of them. You need to know they exist, understand roughly what each one does well, and be willing to try new ones as they emerge. That mindset alone puts you ahead of most people in most workplaces.

AI-assisted writing has become standard practice in 2026 and AI writing skills are growing in demand.

This is not about letting AI write everything for you. It is about knowing how to use AI as a first draft tool, an editing partner, and a thinking aid. The people who do this well are producing more output, with higher quality, in less time than those who do not.

The skill here is knowing when to use AI and when to bring your own voice in. AI is good at structure, first drafts, and summarising. You are good at nuance, judgment, and knowing what your specific audience actually needs. Combine both and you become genuinely fast and genuinely good.

This one is underrated and increasingly important.

AI ethics and governance are now a boardroom conversation and not just a compliance checkbox. AI literacy prevents costly errors.

Knowing the limits of AI tools matters. Understanding that AI can hallucinate facts. That it reflects biases in its training data. That it should not be used for sensitive personal decisions without human review. That outputs need to be verified before being shared externally.

The professionals who understand this are the ones organisations trust with more responsibility. Because they use AI powerfully without creating legal, reputational, or ethical problems for their teams.

If all of this feels like a lot, here is the simplest possible starting point.

Pick one AI tool you use at work, whether that is ChatGPT, Copilot, Gemini, or anything else. For the next two weeks, use it deliberately every single day. Try different prompts. Test different approaches. Notice what works and what does not. That consistent practice is worth more than any course.

AI skills helped offset conventional disadvantages in hiring. Older applicants and candidates without advanced degrees saw their prospects improve substantially when AI skills were present on their resumes.

That is the real story here. AI skills are becoming a leveller. They do not care about your age, your degree, or how long you have been in the industry. They reward curiosity, practice, and a willingness to adapt.

The window to get ahead of this is still open. It just will not stay open forever.

Start today.

— Roo