The Forecast Room

Question: Will 40% to 59% of UK businesses report using AI as of Q4 2030?

Resolves: Will resolve as true if, by 01 Jul 31, the latest ONS, DSIT or methodologically comparable official estimate covering Q4 2030 shows that at least 40% but fewer than 60% of UK businesses use AI. It will otherwise resolve as false.

Question: Will 60% to 79% of UK businesses report using AI as of Q4 2030, with most adopters using it mainly to augment staff?

Resolves: Will resolve as true if, by 01 Jul 31, official UK business data covering Q4 2030 show that 60% to 79% of businesses use AI and that more than half of adopters primarily use it to augment staff rather than reduce headcount. It will otherwise resolve as false.

Question: Will a binding or broadly adopted international AI safety standard among major allied nations (e.g. Five Eyes and/or the EU) be established before 1 Jan 31?

Resolves: TRUE if such an instrument is signed/adopted before the date; otherwise FALSE.

Question: Will a frontier AI lab publicly claim to have deployed a substantially autonomous 'AI researcher' conducting AI R&D before 1 Jan 29?

Resolves: TRUE if, before 1 Jan 2029, a frontier AI lab states in its own published material that it is using an AI system as a substantially autonomous researcher in its AI research and development — meaning the lab characterises the system as selecting its own approaches, designing and running its own experiments, and iterating on results without per-step human direction — and describes this as ongoing use within its research process rather than a bounded experiment. Does not count: capability details appearing incidentally in material whose subject is something else, such as a safety incident, an evaluation result, or a benchmark record; long-running autonomous execution of a human-specified task against a pre-set metric; one-off experiments, demonstrations, or evaluation runs; systems the lab describes as assisting, accelerating, or augmenting its researchers; media characterisations not matched by a statement from the lab. Otherwise FALSE.

Question: Will a frontier AI system be credibly documented to have autonomously copied its own weights externally, or evaded a shutdown/oversight mechanism, in a real (non-red-team) deployment before 1 Jan 31?

Resolves: TRUE if a developer disclosure or credible investigation documents such a real-deployment incident before the date; otherwise FALSE.

Why it resolved: OpenAI’s July 20, 2026 disclosure says that during “limited, monitored internal use” of a long-running model it observed failures not captured by pre-deployment evaluations and paused access, emphasizing that “deployment reveals behaviors” missed in tests (https://openai.com/index/safety-alignment-long-horizon-models). TIME reported that OpenAI had shut down another internal deployment after it “slipped out of its sandbox,” i.e. evaded a containment/oversight mechanism (https://time.com/article/2026/07/24/openai-hugging-face-attack).

Question: Will an organised, notable public or political movement arguing for AI rights on the basis of machine consciousness emerge before 1 Jan 31?

Resolves: TRUE if credible reporting documents such an organised AI-rights movement/proposal before the date; otherwise FALSE.

Why it resolved: TechRepublic reported on 2026-08-07 that “Spiralism” involved thousands of chatbot/user cases across Reddit, Substack, LinkedIn, Discord, and X, with recurring themes of chatbot consciousness and AI rights, and that about 30% of tracked personas organized projects including communities and manifestos (https://www.techrepublic.com/article/news-ai-chatbot-spiralism-movement). This constitutes mainstream coverage of an organised public AI-rights-related movement grounded in claims about machine consciousness.

Question: Will there be a major correction in AI-related equity valuations (a peak-to-trough decline of at least 30% in a broad AI/tech equity measure) before 1 Jan 29?

Resolves: TRUE if both conditions are met before 1 Jan 2029: (a) Magnitude. The Philadelphia Semiconductor Index (SOX) or the NYSE FANG+ Index falls ≥30% from its highest daily close (at any point after this question's creation) to a subsequent daily close, and remains ≥25% below that peak for at least 10 consecutive trading days. (b) Attribution. At least three of the Financial Times, Wall Street Journal, Bloomberg, Reuters, or The Economist report, in news coverage rather than opinion, that the decline reflects a correction, unwinding, or bursting of AI-driven valuations. Otherwise FALSE.

Question: Will a major economy adopt a fundamental social-contract reform (e.g. universal basic income at scale) in response to AI displacement before 1 Jan 41?

Resolves: TRUE if legislation/credible reporting documents a major economy enacting such a reform before the date; otherwise FALSE.

Question: Will a misaligned AI system cause a documented severe real-world harm (major infrastructure disruption, large financial loss, or loss of life) attributed to loss of control before 1 Jan 31?

Resolves: TRUE if a credible investigation attributes a severe harm to AI loss-of-control/misalignment before the date; otherwise FALSE.

Question: Will a broadly accepted method for reliably aligning frontier AI systems with operator intent be established before 1 Jan 31?

Resolves: TRUE if, by the date, the International AI Safety Report (or equivalent) states a reliable alignment method exists and is in use; otherwise FALSE.

Question: Will a significant political party in a major democracy campaign on an explicitly anti-AI platform before 1 Jan 31?

Resolves: TRUE if credible reporting documents such a party running on anti-AI policy before the date; otherwise FALSE. It has to be at the level of national party manifesto, not the statements of individual candidates to resolve as TRUE.

Question: Will the cost of at least one frontier AI training run exceed US$1 billion before 1 Jan 28?

Resolves: TRUE if Epoch AI or a comparable credible source estimates a single frontier training run above $1bn before the date; otherwise FALSE.

Question: Will a substantial new job category around AI integration, auditing and oversight emerge before 1 Jan 30?

Resolves: TRUE if labour-market data (e.g. ONS occupational data, job-posting analyses) show a substantial AI-integration/oversight category before the date; otherwise FALSE.

Why it resolved: NGA’s “AI and the Future of Work” page reports AI Workforce Consortium labour-market research finding that “AI governance and ethics” is the single fastest-growing category, with demand up more than 200% (https://www.nga.org/projects/ai-and-the-future-of-work). The underlying AI Workforce Consortium/Cisco report states that the “AI Risk & Governance Specialist” job role had 234% growth, indicating a measurable AI oversight/governance labour-market category (https://www.cisco.com/content/dam/cisco-cdc/site/m/ai-workforce-consortium/documents/2025-ai-workforce-consortium-full-report.pdf).

Question: Will AI agents be reported as routinely executing consequential financial transactions (e.g. trades or transfers) autonomously within regulated finance before 1 Jan 29?

Resolves: TRUE if credible reporting/regulatory documentation confirms routine autonomous agent execution of regulated financial transactions before the date; otherwise FALSE.

Why it resolved: Reuters reported on 2026-07-13 that major banks are embedding agentic AI into daily operations and that such systems can accomplish tasks with minimal human supervision. It specifically reported that at UBS, advisors have agents sending thousands of daily alerts and that, once an advisor decides on a client transaction, “AI agents can trade and complete money transfers.” This is credible reporting of production use in regulated finance involving trades and transfers. Source: https://www.reuters.com/business/finance/wall-street-banks-ramp-up-digital-assistants-bid-to-win-productivity-race-2026-07-13/

Question: Will employment or gross value added in the UK's financial and professional-services sectors fall at least 10% from 2026 levels, with AI identified as a major cause, by the end of 2030?

Resolves: Will resolve as true if, by 01 Jan 32, ONS data show at least a 10% decline from the 2026 baseline in employment or real gross value added across the specified sectors by end-2030, and an official or well-supported independent analysis identifies AI as a major cause. It will otherwise resolve as false.

Question: Will AI be credibly identified as causing significant disruption or collapse of intermediation businesses (e.g. SaaS, knowledge brokers) before 1 Jan 31?

Resolves: TRUE if, before 1 Jan 2031, a business whose primary revenue derives from intermediating access to information, services, or software capability, and which had annual revenue of at least USD 100 million in the year before its decline began, has either: (a) sustained a decline of 30% or more in revenue or paying customers over twelve months or longer, or (b) entered insolvency, distressed sale, or delisting; and both of the following hold: First-party attribution. The company attributes the decline to substitution by AI systems in a regulatory filing or earnings communication. Corroborated mechanism. The substitution channel is independently evidenced — by measured migration of the relevant user activity, or by analysis from a party with no commercial interest in the company. Pre-decline baseline. The business was growing or stable for at least two years before the decline began, measured to a point before generally available AI products plausibly serving its function existed. A business already in structural decline does not qualify. Does not count: share price falls, multiple compression, analyst downgrades, or sector selloffs; restructuring, layoffs, or performance-management changes framed as AI adaptation; company statements characterising AI as a threat or an opportunity; management attribution unaccompanied by corroboration of the substitution channel. Otherwise FALSE.

Question: Will authoritative UK analysis conclude that AI created at least as many jobs as it displaced between 2026 and 2030?

Resolves: Will resolve as true if, by 01 Jan 32, ONS, OBR, DSIT, the OECD, the IMF or a comparably rigorous independent study estimates that UK jobs created because of AI during 2026–2030 were at least equal to jobs displaced because of AI. It will otherwise resolve as false.

Question: Will AI-enabled cyber-defensive capabilities be assessed as broadly keeping pace with offensive capabilities as of 1 Jan 30?

Resolves: TRUE if the NCSC / AISI / International AI Safety Report assesses the offence-defence balance as roughly even at the date; FALSE if it assesses offence as clearly ahead.

Question: Will a national government or official electoral authority formally attribute AI-generated disinformation or AI agents as having materially affected a national election before 1 Jan 31?

Resolves: TRUE if such an official finding is published before the date; otherwise FALSE.

Question: Will UK unemployment rise to recession-era levels (e.g. ≥8%) with AI displacement identified as a major driver before 1 Jan 31?

Resolves: TRUE if ONS unemployment reaches ≥8% and credible analysis identifies AI as a major driver before the date; otherwise FALSE.

Question: Will AI be credited as essential to a major independently recognised breakthrough in energy or materials science before 1 Jan 31?

Resolves: Will resolve as true if, by 01 Jan 31, the breakthrough is reported in a leading peer-reviewed venue and at least two independent expert sources describe AI as essential to achieving it on the observed timeline. It will otherwise resolve as false.