The safest forecast is often the least useful: technology will advance, policy will respond and outcomes will vary.
Scenario analysis asks a harder question. If several forces already visible develop in different combinations, what kind of society could emerge?
The following futures are not predictions. None assumes a technological breakthrough on a particular date. Each begins with the same 2026 conditions:
- AI adoption is accelerating;
- school attendance and youth participation remain uneven;
- government has set 2035 ambitions for trusted adults and enrichment;
- entry-level work is exposed to task automation but new sectors are growing;
- housing and transport continue to shape independence;
- online regulation is moving towards stronger age controls.
What changes is implementation.
Education, youth services, housing and much of transport are devolved. These scenarios describe broad British patterns, not uniform outcomes across four nations.
Scenario rule. A useful scenario must be plausible, internally consistent and reversible through identifiable choices. It must not be presented with a probability that the evidence cannot support.
Future One: managed reconnection
By 2036, Britain uses technology heavily. It also becomes easier to participate physically.
The National Youth Strategy survives changes of minister and budget. Local authorities receive multi-year settlements. The target for trusted adults is measured through sustained relationships rather than registrations. Free and low-cost activity is linked to schools, youth hubs and transport.
AI enters classrooms under a human-first settlement. Teachers use it for preparation, feedback and targeted support. Pupils learn to challenge outputs, explain reasoning and work with one another. Remote provision supports absence but cannot silently become a permanent substitute for attendance.
Employers receive incentives to preserve training tasks. Foundation apprenticeships expand into a clear first-rung guarantee. Firms that automate junior work contribute to shared training routes, simulated practice and paid placements. AI makes novices productive earlier without removing supervision.
Local transport plans measure access to education, work, activity and town centres in evenings and weekends. Young people help design public space. Libraries, youth centres, sports facilities and community venues operate as a network rather than isolated projects.
AI companions remain available to adults and in restricted forms to young people, but systems must disclose their nature, avoid dependency design and provide clear escalation to human help. Public services retain a reachable human route.
Daily life in 2036
A sixteen-year-old uses an AI tutor to prepare for an engineering foundation apprenticeship. The tool helps identify gaps; a teacher checks understanding. A travel pass reaches the college, placement and youth space. The employer receives support for structured supervision. The young person encounters peers, technicians, customers and adults outside the family.
Technology reduces barriers. Institutions provide consequence, reciprocity and belonging.
Early indicators
- persistent absence continues falling towards the pre-pandemic range;
- under-21 work and apprenticeship entry rises;
- the enrichment participation gap narrows across regions and disability;
- more young people report a sustained trusted-adult relationship;
- youth public-transport use and independent journeys rise;
- AI use grows without corresponding growth in social withdrawal.
What could derail it
Revenue funding may not match capital ambition. Employers may take productivity gains without investing in entry. Age restrictions may prove easier to legislate than enforce. Local capacity may vary. Political attention may move before outcomes mature.
Managed reconnection requires continuity, not one decisive intervention.
Future Two: uneven adaptation
By 2036, national averages look acceptable. Distribution worsens.
AI increases productivity and access. High-performing schools integrate it carefully. Affluent families combine AI tutors with sport, travel, mentors and work experience. Capable employees use AI to extend their judgement and move into higher-value roles.
Elsewhere, AI becomes the cheap layer.
Schools under pressure rely on automated practice. Public services use chat interfaces because staffed contact is expensive. Young people outside education receive digital guidance while
waiting for local help. Remote appointments remain available but human routes become harder to reach.
The youth strategy delivers new buildings and programme places, but participation remains tied to transport, confidence and family capacity. Some cities build strong hubs; rural and financially constrained areas offer thinner provision. A target can be met nationally while particular places fall further behind.
The labour market grows in priority sectors, yet employers recruit fewer true beginners. Graduates and well-connected young people use internships, networks and AI portfolios to enter. Others are told to acquire experience independently.
Daily life in 2036
Two eighteen-year-olds have access to the same AI system.
One uses it between conversations with a teacher, employer and mentor. The other uses it because those people are unavailable. Their technical access is equal. Their human infrastructure is not.
Early indicators
- AI adoption rises faster than supervised training;
- entry-level vacancy requirements increase;
- enrichment targets improve nationally but gaps persist locally;
- digital public-service use rises while satisfaction diverges by income and disability;
- attendance improves for most pupils while severe absence remains concentrated;
- young adults with family wealth achieve independence earlier.
Why this is the most administratively comfortable future
Every programme can show success. AI access expands. Youth facilities open. Employment grows. Average participation improves.
The failure appears between averages.
Uneven adaptation does not look like national collapse. It looks like different classes of human support.
Future Three: accelerated isolation
By 2036, Britain remains connected, productive and socially thinner.
AI becomes the default first contact for education, customer service, benefits, health triage, workplace questions and emotional support. Human help exists but is rationed to complex cases. The threshold for speaking to a person rises.
Employers automate routine junior tasks without creating replacement training routes. Entry-level roles demand immediate competence. More young people remain in education longer, assemble credentials and still struggle to enter work.
School absence becomes managed through sophisticated remote provision. This prevents complete educational loss but weakens the urgency of return. A smaller group of children becomes technically enrolled and socially detached.
Social-media restrictions reduce access to major platforms for under-16s, but offline provision does not expand at the same pace. Use migrates to gaming, messaging, AI characters and less visible services. Parents receive responsibility without practical alternatives.
Housing delivery falls short or remains disconnected from jobs and transport. Young adults spend longer in family homes. Remote work allows employment without widening local networks.
Public space becomes safer, cleaner and more commercial. Teenagers are welcome when participating in organised activity, less welcome when simply present.
Daily life in 2036
A young person can complete lessons, receive career advice, apply for work, speak to a wellbeing assistant, shop and socialise without leaving a bedroom.
Nothing has technically been denied.
Very little requires membership in a shared world.
Early indicators
- severe absence remains elevated despite better remote provision;
- the share of genuine junior roles and under-19 apprenticeships falls;
- AI companions become a common source of primary emotional support;
- human escalation in public services becomes slower or more conditional;
- independent youth travel declines;
- youth-strategy targets measure registrations rather than sustained participation;
- loneliness and in-person contact diverge from measures of digital connection.
Why this future is plausible without conspiracy
Every individual decision is defensible.
Automation saves money. Remote provision preserves access. Age restrictions protect children. Digital services are convenient. Housing takes time. Organised spaces reduce disorder.
The isolation is produced by accumulation.
The decisions separating the futures
AI in education
Does AI return teacher time to relationships, or replace teacher contact? Is remote provision a bridge back or a settled destination?
AI at work
Do employers retain paid entry and supervision, or treat instant productivity as the minimum requirement?
Youth policy
Are targets measured as provision, attendance or sustained relationship? Does revenue funding survive after buildings open?
Digital protection
Are restrictions paired with attractive offline alternatives? Are AI companions regulated for dependency as well as content?
Transport and housing
Are homes and services connected to participation without a car? Does affordability create actual independence?
Public services
Is digital access an option, or the price of receiving help?
The future most likely without deliberate coordination
The evidence cannot assign probabilities.
Institutionally, uneven adaptation is the path of least resistance. Britain already has promising programmes, capable institutions and substantial investment. It also has fragmented delivery, geographic inequality and pressure to use technology for efficiency.
Managed reconnection requires government to protect functions that no department owns naturally: entry, belonging, informal contact and trusted relationships.
Accelerated isolation requires no manifesto. It can emerge if each system optimises its own output.
That is why scenario planning matters.
The purpose is not to announce which future will happen.
It is to identify the indicators early enough to change direction.
2036 will be shaped less by the power of the technology than by the strength of the institutions surrounding its use.
Evidence used across this analysis: [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
Source notes
Evidence cut-off: 17 August 2026. Forecast and scenario limits are identified in the text.
- UK Government, AI Scenarios 2030 (15 June 2026). Open source ↩ Return
- UK Government, Frontier AI capabilities and risks: scenarios annex (28 April 2025). Open source ↩ Return
- DCMS, Youth Matters: Your National Youth Strategy. Open source ↩ Return
- Department for Work and Pensions, Young people and work: interim report. Open source ↩ Return
- Skills England, Annual Skills Report 2026. Open source ↩ Return
- Ofcom, AI chatbots and online regulation. Open source ↩ Return
- Ofcom, Passive social media use, AI companionship and online lives (2 April 2026). Open source ↩ Return
- Department for Education, Pupil absence in schools in England: 2024/25. Open source ↩ Return
- Office for National Statistics, Household projections for England: 2022-based. Open source ↩ Return
- Australian eSafety Commissioner, Social media age restrictions. Open source ↩ Return