In January 2023, ChatGPT was banned by virtually every major school system on earth. By January 2025, most of those same institutions had quietly reversed course. The 39 months between ChatGPT's public release and mid-2026 represent the fastest institutional policy reversal in modern higher education history. The destination is still not settled.
This matters to you not as an education story but as a hiring and management story. The people entering your organisation right now were educated during the most incoherent period of AI policy in academic history. Some learned to use AI well. Some were trained to fear it. Most got conflicting signals depending on which professor they had that semester.
Where Schools Actually Stand
The current state across most major institutions is disclosure plus instructor discretion. The institution sets a baseline, usually that students must disclose AI use, and each professor sets the specifics for their own course. One allows AI for brainstorming but not drafting. The next requires full AI integration. The one after that prohibits it entirely.
A February 2026 analysis of 174 university AI policies found that 117 of them had no explicit AI policy, no disclosure requirement, and no enforcement mechanism. This is 67% of the universities analysed. At the same time, 95% of UK undergraduates are now using generative AI in some form, and 94% are using it for assessed work. The gap between what students are doing and what institutions have a coherent position on is significant.
Source: GradPilot, "Can You Use ChatGPT? AI Policies for 170+ Universities," February 2026. gradpilot.com. Higher Education Policy Institute (HEPI) Student Generative AI Survey 2026. hepi.ac.uk.
What Parents Can Do
If your child is in secondary school now, the questions worth asking their school are specific. Does the school have a written AI use policy, or is it left to individual teachers? Are students being taught how to verify AI output, or just told not to use it? Is AI literacy treated as a skill, the way research skills or citation formatting are taught, or is it treated purely as a disciplinary issue?
The schools handling this well are teaching students to use AI as a tool with visible, documented limitations. They are teaching citation and disclosure. They are running exercises where students deliberately find AI errors and explain why the output failed. That is the curriculum that produces a graduate who is an asset in 2030. The school that bans AI entirely is producing graduates who are inexperienced with a tool they will use every day of their working lives.
For university students already enrolled: check the written AI policy for each course at the start of each semester. Do not assume the institution's general policy covers a specific assignment. The February 2026 analysis found that 56 of 174 universities have at least one programme whose AI policy differs from the institution-level classification. Duke, for example: the Law School prohibits AI while undergraduate admissions permits it.
Source: GradPilot, February 2026. gradpilot.com.
What the Incoherence Produces
The institutions handling this well have three things in common: a written definition of acceptable AI use, a standardised disclosure process, and detection tools that give faculty evidence to act on. The ones handling it badly have each professor setting their own rules, producing graduates who carry conflicting intuitions about what responsible AI use looks like.
The detection tools have their own problem. Turnitin's own guidance says detector scores should be treated as prompts for investigation, not proof of wrongdoing. Students writing in a second language, or with a structured analytical style, can trigger false positives. Several students have faced academic consequences for work they wrote themselves.
The practical result: two graduates from the same institution, same programme, same year, may have radically different AI competencies depending on which professors they happened to take courses with.
Source: Proofademic, "AI and Academic Integrity: A Guide for Institutions," March 2026. proofademic.ai.
What Industry Is Already Doing
The financial services industry is the leading indicator here. Banks are not waiting for universities to sort this out.
JPMorgan's LLM Suite is now deployed to more than 230,000 employees. Jamie Dimon has said publicly that AI "will eliminate jobs" while committing to retraining affected staff. Goldman Sachs CEO David Solomon said in January 2025 that AI can now complete 95% of an S-1 IPO prospectus in minutes, work that once took a six-person team two weeks. His specific observation: "The last 5% now matters because the rest is now a commodity."
In 2025, 88% of AI-related job listings in financial services were for AI user roles, not AI developer roles. Banks are hiring people who can work with AI effectively, not people who can build it. That distinction matters for what hiring managers are actually testing.
Zapier has gone furthest in making this explicit. The company publicly sorts employees into four AI literacy tiers: unacceptable (resistant to AI), capable (basic drafting and summarising), adoptive (integrates AI into daily work), and transformative (builds new workflows around it). Every candidate is measured against that ladder whether they know it or not.
The broader picture: McKinsey research from November 2025 found demand for AI fluency jumped nearly sevenfold in two years, from approximately 1 million workers in 2023 to around 7 million in 2025. LinkedIn's January 2026 labour market report found a 70% year-over-year increase in US roles requiring AI literacy. More than a third of entry-level jobs now explicitly require AI skills, up from roughly 12% in late 2025.
The paradox, identified clearly by Deel's Head of Talent Acquisition: companies desperately need AI literacy, yet interviewers often do not know how to assess it. "You've got motivation to hire an AI-literate workforce, yet there's an immaturity within your existing organisation in terms of the maturity of AI deployment. That creates a situation where the interviewer doesn't really know how to assess the very skill you're trying to hire for."
Source: American Banker, "The Biggest AI Spenders Are Hiring More, Too," June 2026. americanbanker.com. PwC 2026 AI Jobs Barometer. McKinsey, November 2025. Deel, "Companies Want AI Literacy But Can't Assess It," March 2026. deel.com. Interviewpal, "Putting AI Literacy on Your Resume," May 2026. interviewpal.com.
What to Actually Ask in an Interview
The standard questions are a starting point. "Tell me about a time AI gave you something wrong" and "how do you decide when to trust AI output" will surface basic competency. But they are easy to prepare for, and by 2026 most candidates have rehearsed answers.
The more revealing questions go one layer deeper.
Ask them to walk you through the last piece of work they used AI on, step by step. Where exactly did they use it? What prompt did they use? What did they check and how? What would they do differently? A candidate who can narrate that process specifically, not generally, has a working relationship with the tool. A candidate who gives you a principle without a specific example does not.
Ask them about a case where they disagreed with an AI output that looked correct. Not where the AI was obviously wrong, but where the output was plausible and they still chose not to use it. Why? What told them something was off? This question tests calibration, not just error-catching. The dangerous AI user is not the one who ignores AI entirely. It is the one who cannot distinguish between a confident answer and a correct one.
Ask them what AI cannot do in their specific domain. Not in general, in their field, in the kind of work they are applying to do. A candidate who can name specific, concrete failure modes of AI in their discipline has spent real time with the tools in a professional context. A candidate who gives you a generic answer about hallucinations has read about it.
Finally, ask them to show you something. Give them a task during the interview: a short piece of analysis, a draft, a summary of something you hand them. Tell them they can use whatever tools they normally would. Then watch. How they approach a live task with AI visible tells you more than any answer to any question about AI.
Solomon's observation about the last 5% is the right frame for all of this. If AI handles 95% of an S-1, the person who can reliably deliver the last 5% — the judgment, the context, the accountability — is the hire. The interview should be designed to find that person.
Source: Zapier AI literacy framework, zapier.com. Goldman Sachs CEO David Solomon, Bloomberg interview, January 2025. Deel, "Companies Want AI Literacy But Can't Assess It," March 2026. deel.com.
Where Agent Micho Fits
The interview process surfaces the gap. The workflow closes it.
A new hire who is excellent at using AI but has never worked within a structured system will develop habits on the job, some good and some not. A workflow that defines where AI sits in each process — what it produces, what gets checked, what gets escalated to a human — removes the dependency on individual judgment for the parts that should be standardised.
We build those workflows. Not as a replacement for human judgment, but as the structure that makes human judgment show up in the right place at the right time. The last 5% Solomon is describing is judgment and accountability. A well-built system reserves that 5% for the person and handles everything else.
If you are building a team that works with AI, the question worth asking us is not "how do I automate this?" It is "where should a human be in this process, and how do I make sure they show up there reliably?" That is a workflow design question, and it is what we do.