Foundations · Brief 6 of 6

Ten Predictions for AI in 2027

Where the frontier labs are heading, as ten dated predictions with a confidence, the evidence, and what would prove each wrong. The AI Market Map will mark each one right or wrong on its check date.

10predictions
6.4expected true if calibrated
Jun 2027first checks
Brierhow we will score
Home›AI Market Map›Foundations · Brief 6 of 6
FR
FrontierAGI Research
November 18, 2026 · 7 min read · ⚠️ = secondary source
Key takeawaysThe predictionsWhere the labs are less likely to goMethod and sources

⚡ Key takeaways

💡
  1. Most confident (85%): every frontier lab sells its own agent platform by mid-2027.
  2. Restricted tiers for cyber and bio become standard (75%); more labs publish self-improvement ratings (70%).
  3. Prices halve again (70%) while open models stay within six months of the frontier (60%).
  4. Most enterprise agent pilots still stall (65%).
  5. Labs are less likely to certify each other or run inside customers' data centres (55%).

🗺️ The predictions

🎯
Thesis 1: Ten claims, each falsifiable and dated.medium confidence
#PredictionConfidenceCheck byWrong if
1Every frontier lab ships its own agent platform (harness + builder + runtime) to enterprises85%2027-06A frontier lab exits or spins out its agent platform
2Open-weight models reach parity with the best closed model on a major coding benchmark within 6 months of each closed release60%2027-06The gap on the Benchmark Tracker widens two quarters in a row
3At least two more labs publish a self-improvement capability rating for a released model70%2027-06No new self-improvement ratings in system cards by mid-2027
4Third-party evaluation becomes a routine purchase for EU general-purpose AI providers with systemic risk55%2027-12Obligations delayed or evaluations kept fully in-house
5Price per million tokens for frontier-class models falls at least 50% again by end of 202770%2027-12List prices flat or rising across three labs
6Restricted-access models for cyber and bio become a standard product tier at three or more labs75%2027-06Labs fold these back into general models
7Agent-to-agent purchasing (an agent buys a service from another company's agent) appears in a major platform's GA product45%2027-12No GA agent-payment product by end 2027
8Most enterprise agent pilots still do not reach production in 2027 surveys65%2027-12Major surveys report majority in production at scale
9A frontier lab acquires an evaluation, environment or agent-infrastructure start-up60%2027-12No such acquisition announced
10Labs avoid on-premises deployment inside regulated customers' own data centres55%2027-12A frontier lab ships its own on-prem appliance
🔧 Reproduce this: predictions.py
predictions.py · reads data/predictions-2027.json
"""predictions.py — the Interlude's ten predictions for 2027, ready to be scored later.

Reads data/predictions-2027.json. Prints each claim with its confidence and check date,
the expected number that come true if the confidences are honest, and how they will be
scored (Brier score = mean of (confidence - outcome)^2 once outcomes are known).
Standard library only.
"""
import json, os
d = json.load(open(os.path.join(os.path.dirname(os.path.abspath(__file__)), "..", "..", "predictions-2027.json")))
for p in sorted(d["items"], key=lambda p: (p["check"], -p["conf"])):
    print(f"#{p['id']:2d} {p['conf']:.0%} by {p['check']}  {p['claim'][:78]}")
exp = sum(p["conf"] for p in d["items"])
print(f"expected to come true if calibrated: {exp:.1f} of {len(d['items'])}")
done = [p for p in d["items"] if p["outcome"] is not None]
print(f"scored so far: {len(done)}" + (f", Brier {sum((p['conf'] - p['outcome']) ** 2 for p in done) / len(done):.3f}" if done else " (first checks due 2027-06)"))
$ python3 predictions.py # 1 85% by 2027-06 Every frontier lab ships its own agent platform (harness + builder + runtime) # 6 75% by 2027-06 Restricted-access models for cyber and bio become a standard product tier at t # 3 70% by 2027-06 At least two more labs publish a self-improvement capability rating for a rele # 2 60% by 2027-06 Open-weight models reach parity with the best closed model on a major coding b # 5 70% by 2027-12 Price per million tokens for frontier-class models falls at least 50% again by # 8 65% by 2027-12 Most enterprise agent pilots still do not reach production in 2027 surveys # 9 60% by 2027-12 A frontier lab acquires an evaluation, environment or agent-infrastructure sta # 4 55% by 2027-12 Third-party evaluation becomes a routine purchase for EU general-purpose AI pr #10 55% by 2027-12 Labs avoid on-premises deployment inside regulated customers' own data centres # 7 45% by 2027-12 Agent-to-agent purchasing (an agent buys a service from another company's agen expected to come true if calibrated: 6.4 of 10 scored so far: 0 (first checks due 2027-06)
👉
So what: A forecast you cannot be wrong about is not a forecast; these will be scored in public.

📊 Where the labs are less likely to go

📄 Method and sources

⚠️
Facts are sourced; figures marked ⚠️ come from secondary reports or company statements. Scores, weights and predictions are our judgement and are labelled. Disclosure: Claude, made by Anthropic, helps write this site. Anthropic is one of the companies analysed here, under the same rules as the others.

🧭 Foundations: the AI market, Nov 2026

Live data: AI Market Map & Investments · AI Investment Tracker.

6
Predictions for 2027
You are here