University After Intelligence

A thesis on the future of higher education

What is a university for when knowledge is abundant and intelligence is cheap?

For centuries, universities helped organize access to scarce knowledge, expertise, networks, and intellectual development.

Artificial intelligence changes that equation.

The question isn't whether universities still matter. The question is:

What becomes possible now?

By Amir Bomani

AI educator • curriculum designer • AI enablement strategist

The argument in sixty secondsRead the full argument — five parts, 25 minutes

The argument in sixty seconds

Full essay: about 25 minutes

The university should stop competing with artificial intelligence on access to knowledge, and start using it to make human development dramatically better.

  1. 01Knowledge, explanation, and basic intellectual assistance are now abundant. The scarcities the university was designed around have moved.
  2. 02What stays scarce is human: judgment, taste, trust, relationships, accountability, agency, opportunity.
  3. 03So assessment should read capability rather than output, and AI should sit underneath the institution as infrastructure — never on top of it as the product.
  4. 04Better learning is also the strongest institutional argument: capable graduates compound into research, reputation, partnerships, and resources.
Read the full argument (25 min)Jump to the flywheel

Before you disagree

What this thesis claims — and what it does not.

I am claiming

  • The scarcities that shaped the university have changed, so its design should be revisited.
  • Assessment should read capability, not just output.
  • AI belongs underneath the institution as infrastructure, not on top of it as a product.
  • Better learning plausibly strengthens the institution that provides it.

I am not claiming

  • Not that universities are obsolete, or that degrees stop mattering.
  • Not that professors should be replaced, reduced, or automated.
  • Not that every course should use AI — some learning depends on doing it without help.
  • Not that any of this is settled empirical fact. Claims here are labelled by what kind of claim they are.

The thesis

The university shouldn't compete with AI on access to knowledge. It should use AI to make human development dramatically better.

When information and intelligence become abundant, other things become more valuable. Judgment. Curiosity. Taste. Experience. Relationships. Accountability. Community. Discernment. Creativity. Courage. Leadership. Opportunity. The ability to determine what is true. The ability to decide what should be done.

The university of the AI age should increasingly organize itself around developing these capabilities. Not because the old model was irrational, but because it was built for conditions that no longer hold.

Preserve what matters. Reconsider what doesn't. Build what comes next.

Part I

1 / 5

The Collapse

The scarcities the university was built around have stopped being scarce.
I.1

The old bargain

For most of history, intelligence was expensive.

The shape of the modern university is a rational response to five specific scarcities.
  1. Pre-print → 15th c.Books were scarceUniversities concentrated knowledge
  2. 17th–19th c.Experts were scarceUniversities concentrated faculty
  3. 19th–20th c.Research infrastructure was scarceUniversities concentrated laboratories and libraries
  4. 20th c.Professional networks were scarceUniversities concentrated communities
  5. 20th c. → todayCredentials were difficult to obtainUniversities became trusted signals

The industrial learning loop

  1. 01Professor
  2. 02Lecture
  3. 03Reading
  4. 04Assignment
  5. 05Exam
  6. 06Grade
  7. 07Credential

This model wasn't irrational. It was designed for the technological conditions of its time.

I.2

The collapse

Then something changed.

Student

02:13 AM

Intelligence

|

  • RESEARCH
  • EXPLANATION
  • WRITING
  • CODING
  • ANALYSIS
  • TRANSLATION
  • SIMULATION
  • BRAINSTORMING
  • FEEDBACK
  • TUTORING
  • DESIGN
  • PLANNING

A student can now receive personalized intellectual assistance at 2:13 AM. They can ask the same question fifteen times. They can request another explanation. Then another.

They can simulate a debate. Analyze a dataset. Debug code. Explore counterarguments. Translate research. Generate examples. Receive feedback. And continue until something makes sense.

This capability will continue improving.

I.3

The assignment problem

If an assignment can be completed perfectly by AI in 30 seconds, we should question the assignment before we question the student.

This does not mean standards should fall.

It may mean standards can rise.

Old question

“Did the student produce the answer?”

  • Output submitted
  • Format correct
  • Deadline met
  • Score assigned

New question

“Can the student demonstrate understanding, judgment, verification, application, and original decision-making?”

  • 01Understanding demonstrated
  • 02Judgment exercised
  • 03Claims verified
  • 04Knowledge applied
  • 05Decisions owned and defended

Where assessment should move

  1. 01Recall
  2. 02Understanding
  3. 03Application
  4. 04Evaluation
  5. 05Creation
  6. 06Judgment
  7. 07Real-world consequence

AI should push education upward, not eliminate rigor.

Try it — prototype

Redesign one of your own assignments.

Describe something you assign today. See what it could become when intelligence is assumed rather than policed.

Prototype output. Structured for a live model, deterministic for now.

Part II

2 / 5

The New Scarcity

What becomes valuable when knowledge is free — and who develops it.
II.1

The new scarcity

When knowledge becomes abundant, what becomes scarce?

Attention.

  • Judgment.
  • Experience.
  • Trust.
  • Relationships.
  • Accountability.
  • Taste.
  • Courage.
  • Curiosity.
  • Opportunity.

Perhaps the university of the future should organize itself around what remains scarce.

II.2

The product

Maybe the product was never information.

The product is human capability.

A successful university should not merely graduate someone who knows things. It should graduate someone who can learn unfamiliar things, decide what is true, work with people and with machines, and turn knowledge into consequential action.

The human capability stack

AgencyThe capacity to decide what should be done — and to do it.
LeadershipMoving people and institutions toward something better.
CollaborationWorking with people, and with intelligent machines.
CreationMaking something that did not exist before.
JudgmentDeciding under uncertainty; knowing what is true.
ApplicationTurning knowledge into consequential action.
UnderstandingExplaining, connecting, transferring.
KnowledgeNow abundant. Still necessary. No longer sufficient.
II.3

The professor of the future

AI doesn't make great professors less valuable. It makes their most human abilities more valuable.

From

  • Delivering information
  • Grading standardized outputs
  • Repeating explanations
  • Administrative work

To

  • 01Designing experiences
  • 02Challenging assumptions
  • 03Mentoring
  • 04Critiquing
  • 05Facilitating disagreement
  • 06Creating intellectual community
  • 07Connecting theory to reality
  • 08Recognizing potential
  • 09Setting standards
  • 10Helping students develop judgment

The professor becomes less like a search engine and more like a coach, architect, critic, mentor, and intellectual guide.

II.4

The campus

If information moves online, physical community may become more valuable — not less.

  • belonging
  • serendipity
  • friendship
  • debate
  • collaboration
  • identity
  • sports
  • clubs
  • laboratories
  • performance
  • ritual
  • tradition
  • shared experiences
  • mentorship
  • network formation

Universities should not simply digitize themselves. They should ask:

What becomes uniquely valuable when thousands of ambitious humans occupy the same place?

Part III

3 / 5

The AI-Native University

Six design principles, real courses, and the strongest objections to all of it.
III.1

The proposal

Six principles for a university designed around abundance.

Not a university with ChatGPT licenses. A university designed around the reality that intelligence is now abundant.

01

Learn by building

Students should increasingly produce things that exist outside the classroom.

  • Products
  • Research
  • Experiments
  • Campaigns
  • Businesses
  • Software
  • Films
  • Policy proposals
  • Community projects
  • Scientific investigations

The question becomes: what did you make, discover, improve, test, or change?

02

AI becomes infrastructure

Universities don't have “internet departments” responsible for deciding whether students may use Google. Eventually AI may become similarly fundamental.

  • When to use AI
  • When not to use AI
  • How to verify it
  • How to challenge it
  • How to disclose it
  • How to collaborate with it
  • Where human judgment must remain decisive

Infrastructure is taught, governed, and questioned — not merely licensed.

03

Professors become architects of learning

This is not a diminished role. When information is abundant, the professor's scarcest contributions become the most valuable ones.

  • Context
  • Standards
  • Challenge
  • Mentorship
  • Taste
  • Feedback
  • Experience
  • Ethical judgment
  • Intellectual disagreement
  • Domain expertise

Professors increasingly design environments where learning happens rather than functioning primarily as information transmitters.

04

Assess capability, not just output

If AI can produce an essay, evaluate more than the essay.

  • Live defenses
  • Oral examinations
  • Project portfolios
  • Decision logs
  • Process documentation
  • AI interaction histories
  • Peer critique
  • Real-world projects
  • Experiments
  • Demonstrations
  • Reflection
  • Iteration

Output is evidence. Capability is the claim.

What assessment should actually read

01

Output

02

Process

03

Reasoning

04

Judgment

05

Capability

05

Turn the university into a laboratory

Connect coursework to real problems and real counterparts.

  • Local governments
  • Nonprofits
  • Startups
  • Research laboratories
  • Community organizations
  • Small businesses
  • University departments
  • Public institutions

Students learn while producing something useful.

06

Human development becomes central

As machines become more capable, distinctly human development becomes more important.

  • Communication
  • Leadership
  • Collaboration
  • Ethics
  • Resilience
  • Curiosity
  • Taste
  • Empathy
  • Judgment
  • Purpose
  • Agency

This is the part no system can outsource.

III.2

Course lab

What does an AI-native course actually look like?

Theory matters. But redesign becomes clearer when we look at actual classrooms.

Traditional version

Traditional Marketing 301

  • Students read case studies
  • Attend lectures
  • Write marketing plans
  • Take exams

AI-native version

AI-Native Marketing 301

  • 01Every student is assigned a real local business
  • 02Primary research with actual customers
  • 03AI-assisted analysis, human-verified claims
  • 04A campaign that actually launches
  • 05Performance data reviewed with the owner

The traditional loop

  1. Lecture
  2. Reading
  3. Weekly assignment
  4. Essay
  5. Exam
  6. Grade

The AI-native loop

  1. 01Real problem
  2. 02AI-supported research
  3. 03Human verification
  4. 04Build something
  5. 05Receive expert critique
  6. 06Iterate
  7. 07Defend decisions
  8. 08Publish / deploy / present
  9. 09Reflect

Eight weeks, one real business

  1. Week 1

    Interview the owner. Understand the business.

  2. Week 2

    Conduct AI-assisted market research. Verify important claims manually.

  3. Week 3

    Interview customers.

  4. Week 4

    Develop positioning.

  5. Week 5

    Create campaign concepts using AI.

  6. Week 6

    Launch a small campaign.

  7. Week 7

    Analyze actual performance data.

  8. Week 8

    Iterate.

Final assessment

Present the campaign to the business owner and a faculty panel, then defend the decisions behind it.

  • What did you do?
  • Why?
  • Where did AI help?
  • Where was AI wrong?
  • What did you verify?
  • What would you change?

The student leaves with

  • Real experience
  • A portfolio project
  • AI fluency
  • Customer research experience
  • Performance data
  • Expert feedback
  • Professional relationships
  • Demonstrated capability

The assignment becomes harder to fake because the work becomes real.

III.3

Restraint

AI capability does not automatically justify AI use.

Responsible transformation requires identifying where humans should remain central — explicitly, in writing, before adoption rather than after.

Automate

  • Repetitive administrative work
  • Formatting
  • Scheduling
  • Basic information retrieval
  • Initial drafts
  • Routine synthesis

Augment

  • Research
  • Feedback
  • Brainstorming
  • Analysis
  • Curriculum development
  • Student support

Protect

  • Human relationships
  • High-stakes judgment
  • Sensitive conversations
  • Mentorship
  • Ethical decisions
  • Community building
  • Moments where struggle itself creates learning

Good AI strategy is partly knowing where not to use AI.

III.4

Objections

The three strongest arguments against this thesis.

Stated as forcefully as I can state them, before any response. If one of them holds, the argument above needs to change.
01Objection 1 of 3

Thesis statement

AI tutoring makes personalized instruction abundant.

02Objection 2 of 3

Thesis statement

Assess capability through defenses, portfolios, and demonstrations.

03Objection 3 of 3

Thesis statement

AI-native redesign expands what students can accomplish.

An argument that cannot survive its best objection isn't finished being written.

Part IV

4 / 5

The Economics

Why better learning is also the institution's strongest financial argument.
IV.1

The institutional question

There's another question.

What happens to the university itself?

If AI changes how students learn, research, build, create, and work, its effects don't stop at the classroom door. They reach research, employment, alumni outcomes, reputation, partnerships, philanthropy — and eventually the resources available to educate the next generation.

Education is not one system inside the university. It sits at the beginning of an institutional flywheel.

Conceptual modelEverything in this part is labelled by what kind of claim it is.
IV.2

The signature model

The University Progress Flywheel

One node to begin with. Then a system. Then a loop that returns to where it started.
Conceptual modelA proposed institutional mechanism, not measured causation.

AI

Accelerant — not the purpose

Human capability

Who benefits?

This is a loop, not a funnel. Each stage becomes the raw material of the next — and the last stage returns to the first.

Select any stage to see what it produces downstream.

The long form of the same loop

  1. 01
  2. 02More capable students
  3. 03Better projects
  4. 04Better research
  5. 05More discoveries
  6. 06More startups + innovation
  7. 07Better employment outcomes
  8. 08More successful alumni
  9. 09Stronger employer relationships
  10. 10Greater reputation
  11. 11More applicants
  12. 12More partnerships + philanthropy + research funding
  13. 13

Output becomes input

This is a loop. Not a funnel.

Now add AI to the system

AI can accelerate the flywheel.

Not as the centre of the institution. As infrastructure underneath it — touching learning, research, building, analysis, administration, career preparation, and entrepreneurship.

Two lenses on the same loop

Where AI touches the loop — as infrastructure underneath the institution, not as its centre.

AI

  • Learning
  • Research
  • Building
  • Analysis
  • Administration
  • Career preparation
  • Entrepreneurship

AI + LEARNING

  1. 01Student
  2. 02AI tutor
  3. 03Faster feedback
  4. 04More iteration
  5. 05Deeper understanding
  6. 06More capable student

The professor still sets the standard the iteration aims at.

AI + RESEARCH

  1. 01Research question
  2. 02AI-assisted literature exploration
  3. 03Data analysis
  4. 04More hypotheses tested
  5. 05Researcher verification
  6. 06Potential discovery

AI does not produce better research. Verification and domain expertise do.

AI + ENTREPRENEURSHIP

  1. 01Idea
  2. 02Research
  3. 03Prototype
  4. 04Customer testing
  5. 05Iteration
  6. 06Launch

What changes is the cost and time between stages — not the need for judgment.

AI + CAREER READINESS

  1. 01Student
  2. 02AI-native coursework
  3. 03Real projects
  4. 04Portfolio
  5. 05Demonstrated capability
  6. 06Employer

Evidence is produced along the way, not manufactured at the end.

No verified citation attached yet. AI's effect on learning and research outcomes is an open empirical question. These are proposed mechanisms.

The opportunity isn't to make education easier. It's to increase what students are capable of accomplishing during the same four years.

AI isn't the product. Human capability is the product. Progress is the outcome.

IV.3

The bundle

The value proposition is changing.

What exactly is a student paying a university for when world-class explanations are increasingly available on demand?

The traditional value bundle

  • Knowledge→ increasingly abundant
  • Basic instruction→ increasingly abundant
  • Credential
  • Networkstill scarce
  • Communitystill scarce
  • Experiencestill scarce
  • Mentorshipstill scarce
  • Researchstill scarce
  • Opportunitystill scarce
  • Transformationstill scarce

Nothing here becomes worthless. Some parts simply stop being scarce.

As some parts of the university bundle become less scarce, universities have an opportunity to make the remaining parts dramatically more valuable.

From information to transformation

University as information system

  1. Professor

  2. Information

  3. Student

  4. Exam

  5. Credential

University as transformation system

  1. Student

  2. Knowledge + AI

  3. Practice

  4. Mentorship

  5. Real problems

  6. Collaboration

  7. Feedback

  8. Iteration

  9. Experience

  10. Demonstrated capability

  11. Opportunity

The university doesn't disappear. Its job gets bigger.

IV.4

Credential → evidence

What if a degree came with evidence?

A degree should represent more than completed coursework. It should represent demonstrated capability.
Conceptual modelConceptual example. Not actual student data.

Traditional credential

B.A. Marketing

GPA: 3.6

Graduated: 2030

What does this actually tell an employer?

Credential + evidence

Marketing — B.A.

During four years this student:

0
Real-world projects
0
Organizations served
0
Research projects
0
AI systems built
0
Startup launched
0
Public presentations
Credential→Credential + evidence→Demonstrated capability

Student

Instead of

“I took Marketing 301.”

They gain

“I helped three organizations solve actual marketing problems — here's the evidence.”

Employer

Instead of

Degree · GPA · Interview

They gain

Portfolio, projects, recommendations, demonstrated skills, decision-making evidence, real experience

  • Portfolio
  • Projects
  • Recommendations
  • Demonstrated skills
  • Decision-making evidence
  • Real experience

University

Instead of

Completion data

They gain

Evidence that the educational model produces outcomes

  • Its students can perform
  • Its model produces outcomes
  • Its graduates are valuable
  • Its programs connect learning to reality
IV.5

Three reinforcing loops

Capability doesn't stay inside the classroom.

Three mechanisms through which better learning could plausibly strengthen the institution around it.

The employer trust loop

  1. 01University
  2. 02Student
  3. 03Real project
  4. 04Employer
  5. 05Successful hire
  6. 06Employer trust
  7. 07More recruiting
  8. 08More student opportunity

Output becomes input

If employers repeatedly discover that graduates from an institution can actually perform, the institution's credential may become more valuable.

HypothesisA proposed institutional mechanism, not established causal research.

No verified citation attached yet. Treat this as a proposed mechanism.

The research flywheel

  1. 01Faculty + students
  2. 02Research questions
  3. 03AI-assisted exploration
  4. 04Human verification
  5. 05Experimentation
  6. 06Discovery
  7. 07Publication / product / patent / public impact
  8. 08Reputation
  9. 09Funding + talent

Output becomes input

Speed without rigor is not progress.

The goal is not more research output at any cost. The goal is increasing researchers' ability to explore, test, analyze, and discover while preserving rigor.

What happens when students can build sooner?

The entrepreneurship loop

  1. 01Student
  2. 02Problem
  3. 03AI-assisted research
  4. 04Prototype
  5. 05Customer feedback
  6. 06Iteration
  7. 07Product
  8. 08Company / project / organization
  9. 09Jobs + impact + alumni success
  10. 10University ecosystem

Output becomes input

Student idea→Software prototype
Research discovery→Startup
Class project→Nonprofit
Design project→Product
Engineering project→Patent
Journalism project→Publication

Not every student needs to become an entrepreneur. The point is that AI can lower the distance between an idea and an experiment.

IV.6

Who benefits

Why student success matters to the university — and to everyone around it.

These outcomes are not independent. Each one raises the ceiling of the others.

Universities don't need to ask society to trust them.

They can give society reasons to.

The desired student experience

That experience changed what I'm capable of.

The desired employer experience

Their graduates can actually do things.

The desired researcher experience

This environment allows us to discover.

The desired community experience

This institution makes our region stronger.

The desired founder experience

This university helped me build.

The desired alumni experience

My university continues creating opportunities for me.

Conceptual outcomes — described, not quoted. No one said these things yet.

The trust flywheel

  1. 01Better education
  2. 02Demonstrated capability
  3. 03Better student outcomes
  4. 04Employer confidence
  5. 05Alumni success
  6. 06Research + public impact
  7. 07Institutional reputation
  8. 08Greater trust
  9. 09More opportunity

Output becomes input

At the centre

Trust

Trust is not a branding strategy. It is the accumulated result of demonstrated value.

IV.7

The economic argument

Universities are institutions. And institutions need resources.

  • Great faculty require resources
  • Research requires resources
  • Laboratories require resources
  • Student support requires resources
  • Scholarships require resources
  • Technology requires resources
  • Community programs require resources
  • Innovation requires resources

Financial sustainability and educational mission should therefore not automatically be treated as opposing ideas.

A stronger educational model can strengthen the institution that provides it.

The resource loop

  1. 01Better education
  2. 02Better outcomes
  3. 03Stronger relationships
  4. 04Greater institutional value
  5. 05More resources

Output becomes input

Hypothesis“Can,” not “will.” Educational improvements do not automatically increase institutional revenue.

No verified citation attached yet. Treat this as a proposed mechanism.

But there is a dangerous version of this idea.

Faster

  • More output.
  • More content.
  • More applications.
  • More research.
  • More companies.
  • More productivity.

More isn't automatically better.

An AI-native university cannot simply become a productivity factory. Universities also exist to create space for:

  • reflection
  • deep thinking
  • intellectual exploration
  • arts
  • philosophy
  • fundamental research
  • citizenship
  • community
  • relationships
  • ideas without immediate commercial value
  • human development

Efficiency should create more room for humanity — not consume it.

IV.8

The new equation

The university equation, rewritten.

University=Knowledge•Instruction•Credential

What if the university became the best environment in the world for turning human potential into human capability?

So what do we do Monday?

Part V

5 / 5

What To Do Monday

Three first moves, a six-stage method, and the questions still open.
V.1

Monday morning

Three things a department could start this week.

No budget request, no committee, no institutional permission. Each one produces evidence you can put in front of colleagues.

Move 01

Audit ten assignments against one question

One department chair, one afternoon

  1. Run each assignment through a current model exactly as a student would.
  2. Sort results into three piles: AI does this perfectly, partially, or not at all.
  3. The first pile is your redesign queue. Nothing else needs to change yet.

A ranked list of assignments that no longer measure what they claim to measure.

Move 02

Write the restraint list before the policy

Faculty senate or teaching committee, one session

  1. Name the places where human judgment must stay decisive — in writing, in advance.
  2. Name the productive struggle you are deliberately protecting from assistance.
  3. Publish both lists to students alongside whatever AI use you permit.

A short document that makes disclosure and integrity legible instead of adversarial.

Move 03

Run one thirty-day experiment with real measurement

One course, one instructor, one willing cohort

  1. Pick a single course and redesign one assignment sequence, not the syllabus.
  2. Decide what evidence you will collect before you start: time, quality, student experience.
  3. Report the result honestly at the end — including if it did not work.

Evidence you own, from your own institution, that a faculty meeting can argue with.

Start with one thing. Measure it honestly. Then argue.

V.2

Framework

Philosophy without implementation is just a thought experiment.

The AI-Native Institution Framework — six stages, in order.

DISCOVER

Understand how the institution actually operates.

  • Interview students, faculty, administrators, department leaders, and technology teams
  • Map workflows as they exist, not as they are documented
  • Identify frustrations
  • Understand culture

A thirty-day version

  1. Week 1

    Observe + interview

    Sit in. Ask. Document reality.

  2. Week 2

    Map + redesign

    Current state, opportunity map, new design.

  3. Week 3

    Build + train

    Prototype tools, materials, enablement.

  4. Week 4

    Run + measure

    Ship it live. Instrument it. Report honestly.

V.3

Research

This thesis should be challenged.

The future of higher education is too important to build from intuition alone.

AI + Learning

[1]

Under what conditions does AI assistance increase durable learning, and when does it substitute for the cognitive work that produces it?

The entire redesign argument depends on distinguishing assistance that builds capability from assistance that bypasses it.

Research source to be added

Assessment

[2]

How reliably can oral defenses, decision logs, and process evidence discriminate genuine capability from AI-produced output?

If capability-based assessment is not reliable at scale, institutions cannot responsibly adopt it.

Research source to be added

Cognitive Science

[3]

Which forms of desirable difficulty lose their benefit when an external system can remove the difficulty on demand?

This determines which exercises should remain deliberately AI-free.

Research source to be added

Future of Work

[4]

Which graduate capabilities are employers actually unable to obtain from AI systems today?

Curriculum priorities should follow evidence about complementarity, not speculation.

Research source to be added

University Innovation

[5]

What distinguishes institutional pilots that scale from pilots that remain isolated experiments?

The one-department experiment is only useful if there is a path to diffusion.

Research source to be added

AI Literacy

[6]

What does verified AI literacy consist of, and can it be measured independently of tool familiarity?

Teaching AI literacy requires a definition that outlives specific products.

Research source to be added

Human-AI Collaboration

[7]

How does over-reliance form, and which interaction patterns preserve independent judgment?

The failure mode of an AI-native university is quiet intellectual dependence.

Research source to be added

Learning Science

[8]

How do project-based, real-consequence courses compare with conventional courses on transfer and retention?

Learn-by-building is a claim about outcomes, and should be tested as one.

Research source to be added

Institutional Trust

[9]

What does a credential signal to employers and the public once AI-assisted work is normal?

Trust in the credential is the university's most valuable asset.

Research source to be added

No studies, statistics, authors, or quotations are cited on this site until they have been verified. The citation infrastructure exists; the shelf is deliberately empty of unverified claims.

Open questions

Questions I don't think we have answered yet.

  1. 01What happens to learning when answers become effortless?
  2. 02Which forms of productive struggle should education preserve?
  3. 03How should AI usage be disclosed?
  4. 04What does academic integrity mean when AI becomes infrastructure?
  5. 05What should students memorize?
  6. 06How do we prevent AI from weakening independent thinking?
  7. 07How should universities assess AI-assisted work?
  8. 08How do we preserve intellectual disagreement?
  9. 09What happens when AI tutors become better than average human tutoring?
  10. 10Which human skills increase in value?
  11. 11What does a meaningful credential measure?
  12. 12What should four years of university actually produce?
  13. 13How do we ensure AI expands opportunity rather than inequality?

These questions deserve experiments, not slogans.

V.4

Work with me

Give me one thing worth redesigning.

A course. An assignment. A department. A faculty workflow. A student experience. A recurring administrative problem. Let's find out what it becomes in an AI-native university.

I work at the intersection of AI, education, curriculum, systems, technology, and creative problem solving. Former AI Specialist inside the GaryVee / Vayner ecosystem. Ten-plus years teaching and coaching, including teaching computer science. Today: AI consulting and enablement, building AI workflows and products, and developing curricula and workshops.

My interest isn't teaching people to use AI tools. It's what happens when we redesign systems around capabilities that didn't exist when those systems were created.

LinkedIn Proof of work