How to AI

 

Axios Finish Line: How to AI

• Jim VandeHei,
• Dan Cox

Illustration: Brendan Lynch/Axios

Axios CEO Jim VandeHei and Axios chief technology officer Dan Cox were chatting about how to share the basics of AI, and this prompting primer was born:

The No. 1 pushback we get from AI skeptics or newbies is: "It's overhyped. I asked it something and it spit out an unimpressive answer!"

• Truth bomb: It's not the AI. It's you.

Why it matters: Ask top large language models like ChatGPT something simple or generic, and you will get a simple or generic answer. Ask the right questions the right way, and you will often get magic.

Jim asked Dan Cox, our CTO and AI leader, to help him craft six ways for ordinary users to get more extraordinary answers. It starts with prompts — the very questions you pose to ChatGPT, Claude, Gemini, Grok or Perplexity.

• Just in time for performance-review season, here are six tips (for both reviewers & reviewees):

1. It's a conversation, not a search engine. The biggest mistake newbies make is treating AI like Google — one question, one answer, done. The magic happens in the back-and-forth. Ask a question. Read the answer. Then push: "Make it shorter ... Give me three alternatives ... That's too formal ... What am I missing?" The best outputs come from the fifth or sixth exchange.

• Example: You ask for help requesting a raise. The first draft is generic. You say: "Too corporate. I've been here six years and my boss is informal — make it sound like a real conversation." Now it's useful.

2. Nail the Who. Start by explaining who you are — your role, experience, anything relevant — and who you want the AI to think like when answering. Be specific.

• Example: "I'm a senior account manager at a midsized software company. I've been here six years, consistently hit my numbers, and just took on two new direct reports. I want you to think like a brutally honest executive coach who has helped hundreds of people negotiate compensation."

3. Context matters most. Give detailed, real-world framing upfront. The AI doesn't know your situation, your audience, or what you're trying to avoid unless you tell it. Specificity is everything.

• Example (building on the Who): "My annual review is in two weeks. I haven't had a raise in 18 months despite a promotion in title. I know the company had a rough Q3, but my division exceeded targets. My boss is supportive, but not the final decision-maker — he has to pitch it to the VP. What's the smartest approach?"

4. Just say no. Tell it what not to do. This sharpens output dramatically.

• Example (adding constraints): "Don't give me generic advice like 'know your worth.' Don't suggest ultimatums — I'm not bluffing. And don't make it sound like a script I'd read verbatim."

5. Say: "Think step by step." When you're dealing with anything complex — a negotiation, a decision with trade-offs, a strategy with multiple variables — ask the AI to reason through it explicitly. This simple phrase dramatically improves output quality.

• Example: "Think step by step about how my boss will react and what objections he might raise when pitching this to the VP."

6. Just dump the image in. The models are extraordinary at instantly understanding screenshots, documents or files. Stop wasting time explaining what you're looking at — the AI can just see it.

• Example: Screenshot your company's salary bands from the internal HR portal. Paste it into ChatGPT alongside your title and tenure. Ask: "Based on this, where should I be? What's a reasonable ask?"

⚠️ Trust but verify. AI can hallucinate confidently — inventing facts, statistics, even citations that don't exist. The more specific the claim, the more you should double-check. Use it to think, draft and strategize. But if it spits out a number or a name, verify it before you repeat it.

The bottom line: AI is a power tool. It rewards users who treat it like a sharp colleague rather than a magic box. Be specific. Be demanding. Keep pushing.

• How'd you do? How'd we do? What's your power prompt? Let us know: finishline@axios.com.

Go deeper: Jim's video, "Blunt AI advice."

America's next class war: AI fluency

 

Behind the Curtain — America's next class war: AI fluency

• Jim VandeHei,
• Mike Allenemail (opens in new window)

Illustration: Aïda Amer/Axios. Stock: Getty Images

Anthropic just dropped the most granular data yet on who's actually using AI and how — and the findings should rattle anyone thinking AI's gains will be evenly distributed.

• It won't. In fact, it's creating a new form of economic inequality: AI fluency.

Why it matters: The Anthropic data, out Tuesday, reveals something subtler and more consequential than the "robots take your job" narrative.

• The real divide isn't between people who use AI and people who don't. It's between experienced AI users and newcomers to AI.

AI continues to pose a serious risk to any automatable jobs, which Anthropic CEO Dario Amodei has warned could wipe out half of entry-level white-collar work.

• Two workflow categories doubled in prevalence between November and February: automated sales and outreach, and automated trading.
• That means people doing those tasks are in more danger of seeing that part of their job go away.

But AI will also be a growing threat to casual or unsophisticated users who fall behind their more AI-savvy peers, regardless of role or level.

• "Much of the discussion focuses on how AI is something that happens to you," Peter McCrory, Anthropic's head of economics, told us from the company's headquarters in San Francisco.
• "This analysis shows you can develop skills that make you better at getting value out of Claude or whatever large language model you want to use."

Some context: Anthropic's new report, "Anthropic Economic Index: Learning Curves," studied over 1 million conversations on the company's Claude platform last month. The headline finding: Experienced AI users get better results out of an AI model than newcomers. And the gap isn't explained by what tasks they're doing, what country they're in, or what model they're using.

• People who've used Claude for six months or more have a 10% higher success rate in their conversations with AI. "The longer you've been using it, the stronger this effect," McCrory says.
• Adoption of Claude in hypereducated Washington, D.C., is four times the adoption rate you'd expect for a city of its size.
• Globally, inequality in usage has persisted since Anthropic's last report, in January, in the 20 higher-income countries with the most Claude use.

That's a skills gap hardening into a class gap in real time. But you can escape it by experimenting, getting comfortable, getting deft, getting fluent.

• Anthropic's researchers are candid that this could be early-adopter selection bias or survivorship — maybe sophisticated users simply signed up first.
• But Anthropic's finding certainly mirrors our personal experience.

Between the lines: People think of AI as a tool, when you should think of it as a never-before-imagined toolbox — it allows you to not just automate a boring task, but stretch your abilities across most things you touch at work. But only once you start to master prompts, and pushback, and persistence when unsatisfying or unilluminating answers come back.

• Jim started using the models like most — like a search engine. But then they became his best researcher ... then idea stress-tester ... then builder of prototypes for new businesses. He's basically at the six-month mark Anthropic describes, and discovering new use cases every week.

You have to move up the AI proficiency ladder. Using a large language model as a search engine or copy editor is dumb AI. Even having it draft emails for you is like having a celebrity chef boil your water.

• The report divides tasks into "automation" (do this task) and "augmentation" — more polished, sophisticated inputs like using the LLM as a thought partner that spits out ideas and feedback, or writes a business plan, or stress-tests a business plan, or coaches and teaches you.
• Think how much more valuable AI dexterity will make you to your current organization — or how much more marketable it'll make you to a future employer.

The big picture: This report lands in the middle of the most anxious era Americans have experienced about AI and jobs since OpenAI's ChatGPT moment after the model's release in late 2022.

• An NBC News poll from earlier this month found that 57% of registered voters believe AI's risks outweigh its benefits.
• Only 26% have positive feelings about the technology — a net favorability lower than that of any other topic polled, except the Democratic Party and Iran. (AI was two points less popular than ICE.)

AI users are getting better, while AI anxiety surges and the job market deteriorates. It's a reality that Washington isn't confronting with consistency and seriousness.

• Washington is debating AI in the abstract: Should we regulate it, should we race China, should we worry about superintelligence?
• But the Anthropic report makes the near-term problem concrete: Signs of a two-tier workforce are already emerging. And neither party has a plan for people on the wrong side of it.
• What Anthropic found in observing real-world use: Skilled AI users are getting better at collaborating with Claude to do a wide variety of work, not just automate specific activities.

The bottom line: The people already using AI for high-value work may pull further ahead, with real implications for who captures the economic benefits of this technology.

• If you're not an early adopter, today's your chance.

OLIVIA OF TROYE

 

Saturday Covfefe: Watch the Pattern

Who pays. Who rises. Who gets pushed out.

Olivia of Troye Apr 4

I spent this week watching the same thing play out across very different stories:

Decisions at the top. Consequences everywhere else.

From veterans on the brink of losing their homes, to shifts inside the military, to a workforce dividing in real time—this isn’t abstract.

Different headlines, same throughline.

1. Veterans Are Losing Their Homes, It Didn’t Have to Happen

More than 10,000 veterans have already lost their homes. Another 90,000 are on the brink, and it didn’t have to happen. This wasn’t complicated, and it wasn’t unforeseen.

The Trump administration shut down a Veterans Affairs rescue program that was actively helping veterans stay in their homes, after being warned that ending it without a replacement would lead to mass foreclosures.

They did it anyway. There was no transition or backup plan, just a bureaucratic cliff.

Veterans who fell behind, many because of injuries, job loss, or delays in disability payments, were left with impossible choices: Pay everything back at once, accept higher monthly payments they couldn’t afford, or lose their homes. Some were literally told not to make payments while waiting for help, and then pushed into foreclosure for falling behind.

These are combat veterans—families relying on a system that’s supposed to provide one of the most meaningful benefits of service—a pathway to stability, to the middle class, to a home. Right now, it’s doing the opposite.

This isn’t just a housing crisis. It’s a policy failure.

Yes, I’m angry. Once again, the people paying the price are those who serve.

🏠 Veterans are losing their homes: NPR

2. The Pentagon Purge Continues

Another one, gone.

Defense Secretary Pete Hegseth just forced Army Chief Randy George to step down early, without explanation. There was no scandal, failure, or reason given. Just removed.

And it’s not happening in a vacuum.

This comes in the middle of the largest U.S. conflict in the Middle East in two decades. And he’s not alone. Top military leaders across the services have been pushed out across the Air Force, Navy, Joint Chiefs, senior intelligence, and even military lawyers.

Let’s call this what it is: A purge.

Because when experienced, independent leadership gets replaced all at once, it’s not about performance. It’s about control.

The military is meant to operate on one principle: mission, law, and the Constitution. Not politics, not loyalty tests, not who aligns with the agenda of the moment. Yet, that’s where we’re headed. The issue is both structural and ideological.

Hegseth has framed this war effort in explicitly religious terms, calling for prayers for victory “in the name of Jesus Christ.” Even Pope Leo XIV pushed back this week, warning that domination and destruction are “entirely foreign to the way of Jesus Christ.” That contrast says a lot.

You don’t remove this many senior leaders this fast, especially in the middle of a conflict, unless you’re trying to reshape the institution itself.

The question isn’t just who’s being pushed out, but who’s being put in, and why. The answer is clear.

🪖 Top Army Chief Ousted: Politico

3. The Real Divide Isn’t AI, It’s Who Knows How to Use It

The divide isn’t coming. It’s already here.

New data from Anthropic shows something more subtle, and more important, than job loss when it comes to Artificial Intelligence (AI). The gap isn’t between people who use AI and people who don’t. It’s between people who know how to use it well and everyone else.

Same tools. Same access. Different outcomes.

People with just a few months of experience are already producing better work—faster, sharper, more effective. Not because they’re smarter. Because they’ve learned how to work with it. That’s the shift.

AI isn’t replacing the workforce. It’s dividing it. A two-tier system is forming in real time: Those who can collaborate with AI, and those who can’t. And yes, the gap compounds.

The more fluent you are, the faster you improve. The further ahead you get.

But here’s the part people are missing: This isn’t fixed.

The barrier to entry is still low. You don’t need permission or a new degree. You just need to start utilizing.

Use it to think, not just search. Push it. Question it. Refine it. Treat it like a partner, not a shortcut.

The real risk isn’t AI. It’s standing still while others learn to move with it.

💡 Want to Stay Ahead? Start Here:

• Don’t just ask—challenge it.

• Don’t settle—refine it.

• Don’t dabble—use it every day.

🤖 The New Class War: Who Knows AI, and Who Doesn’t: Axios

4. Fired in the Limo

This is how it happened. Attorney General Pam Bondi was in "The Beast" (the presidential limo, as it’s often referred to) on the way to the Supreme Court when Trump told her she was out. "I think it’s time."

Then she still had to sit through the hearing on birthright citizenship, side by side with him. Knowing.

Call it what it is, humiliation.

After months of trying to deliver—pushing investigations, chasing targets, taking the heat from the Epstein fallout—it still wasn’t enough.

Earlier this week, a seasoned Army general was pushed out. Now, a loyalist attorney general is gone too. Different people. Different approaches. Same outcome.

So what actually protects you? Not competence. Not loyalty. In this system, neither is enough. This pattern was obvious in Trump 1.0. The real question is why anyone in this latest Cabinet thought they’d be the exception.

The rules shift. The pressure flows down. And anyone can be next, no matter how far they go to cover for Trump and prove themselves.

My suggestion for Pam Bondi:

Olivia of Troye 2d Pam Bondi— you know what’s in those Epstein files. No more hiding. Go public. ALL OF IT. 3.9K 624

The details on how this all went down make it quite the read.

🔥 Bondi fired mid-ride: WSJ

5. The World Is Sending the Bill for Trump’s War

This is where it lands.

Not in speeches. Not in press briefings. In people’s daily lives.

A doctor in New Zealand just sent the U.S. embassy a nearly $3,000 bill for rising fuel costs tied to the Iran conflict. Not as a stunt. As a statement.

After Trump’s Iran speech, oil prices surged, jumping as much as 8–11% overnight. That spike didn’t stay on trading floors. It hit gas pumps worldwide. The doctor’s staff can’t afford to get to work. So the clinic covered it. Now he’s asking the U.S. to pay it back.

Because from where the rest of the world is sitting, this wasn’t inevitable. It was the result of decisions, escalation, and a war that didn’t have to happen.

And while Washington debates strategy… or the lack of one…

The impact is already here:

Higher fuel.

Higher costs.

Real strain on real people.

This invoice isn’t about the money. It’s about accountability.

A reminder that when America escalates, the world pays.

⛽️ The World Is Sending the Bill for Trump’s War: The Guardian

If this helps you see the pattern behind the headlines, consider supporting the work—or buying me a cafecito ☕️.

💫 One Thing For Your Soul

Sometimes the universe course-corrects.

On March 1, Jess McClain was leading the U.S. Half Marathon Championships in Atlanta—commanding lead, world championship spot basically in hand—when the official pace car led her and two other runners off the course with barely a mile to go.

She had to come to a complete stop, make a tight U-turn, and run back onto the course as a national title and a Team USA spot slipped away.

She finished ninth.

Appeals were filed. Appeals were denied. Prize money was offered as a consolation.

But this week, World Athletics granted USA Track & Field a one-time exception to send seven women, rather than the standard four, to the World Road Running Championships in Copenhagen this September. McClain, Hurley, and Kurgat got their spots.

And the woman who officially won? Molly Born made clear she didn’t consider herself the true winner, and said she would likely have turned down a solo team spot she didn’t feel she earned.

Everyone did the right thing. Sports bureaucracy, the official winner, the governing body—all of them.

Proof that ethics and integrity > ego can actually happen.

🏃‍♀️ When Everyone Chooses Integrity: NPR

Happy Passover, Happy Easter, and Happy Spring!

Grateful you’re here.

Time to go prepare for my grand appearance tomorrow at the neighborhood egg hunt…shhh!!!

Until next time,
Olivia