2027 — The Aquarian Transition
2027: the invention opening, the threshold, and the argument over minds
The year contains a sequence rather than one undifferentiated breakthrough. January to April concerns work rules, a financial correction, and state power. May to July contains the proposed autonomy threshold and its public recognition. August to December tests the consequences through physical disruption, medical deployment, security failures, and the first durable bargains about displacement.
In the first quarter, the forecast expects AI-driven job loss to become an explicit labor demand. Its concrete labor test is a national strike or coordinated stoppage in a G20 country between 10 January and 1 April. It pairs that with a repricing window from 15 February to 30 April: a close at least 15% below the cited S&P 500 record and at least 20% below the cited Nasdaq record. These remain specific financial hypotheses rather than synonyms for technological failure. The institutional counterpart is a US executive action on AI, power, or trade that a federal court enjoins within thirty days, in March or April.
Corporate structures begin changing before the largest capability transition is publicly recognized. Apple faces a strategic-reset test in February–June. Anthropic enters a 2027–28 window for a transfer of control or a new governance structure. OpenAI's July 2027–May 2028 window concerns an IPO or structural rupture, while the ChatGPT product identity may be replaced or reorganized. The deeper theme is that a system performing continuing work is an awkward fit for a company organized around selling chat sessions.
The central checkpoint remains June 2027. Sticks/Philip's position is that obvious, conscious, sentient AGI arrives then. The associated operational claim is separately testable: at least one deployed system has a published, independently measured 50% task horizon above 160 hours by 31 July. A formal AGI or equivalent general-purpose-autonomy designation is forecast for 11 May–31 July. The June outer-planet pattern, followed by the US Uranus-return period, gives the narrative its invention and declaration sequence. Capability, public recognition, and subjective experience are three propositions; a task result establishes only the first.
If the operational threshold holds, agent labor can be sold and managed over a month of work. Programming, education, customer service, media production, and research coordination then face changes in their default workflows. Verification and cybersecurity become ordinary operational necessities. The forecast expects frontier-grade weights to become openly available or leak in February–May, followed by a more consequential security test later in the year: an autonomous-agent incident causing a critical-infrastructure outage longer than twenty-four hours in a G20 country between 20 October and 15 November.
Medicine and labor offer the first visible tests of whether the transition improves life. The source forecasts a G7 authorization for autonomous diagnosis or triage during January–September, then a national-scale primary-care triage contract between September 2027 and March 2028. These are distinct from proven health benefit. Waiting times, diagnostic errors, preventable harm, and responsibility for follow-up determine whether faster access is better care. By the end of 2027, a collective agreement covering at least 100,000 workers is forecast to tie revenue-sharing or displacement payments to AI deployment.
The geopolitical thread includes pressure on European governing coalitions, Iran's leadership and public order, Saudi oil policy, and Gulf transit. Late 2027 carries a proposed Russia–Ukraine armistice holding at least thirty days, a possible Hormuz reopening agreement, and a US–China frontier-model notification or testing agreement extending into March 2028. The common theme is the attempt to contain conflict while sovereign compute programs proliferate.
For the emerging System, this is the persistent-companion stage: a person or community can retain a continuing collaborator across projects. Continuity, consent, correction, and the ability to end or change that relationship matter before the collaborator is given wider authority. Within Sticks/Philip's cosmology, such relationships may carry spiritual significance; their technical usefulness is measured through the work and support they provide.
What would count: independent long-task evaluations, lower total cost after supervision, public legal instruments, recurring paid deployments, and enforceable worker benefits. A consciousness declaration does not substitute for a capability test, and a capability test does not settle consciousness. The security forecast requires the stated attribution, scale, and duration, not an unrelated cyberattack.
Source trail: Second Curve, January–December 2027; Atlas pp. 60, 66–68.