The authors of AI 2027 predict two horrifying possible futures. The Bible insists neither is the final word.
Do androids dream of electric sheep?” This question is also the title of Phillip K Dick’s landmark 1968 dystopian sci-fi novel that follows the story of Rick Deckard, a bounty hunter tasked with “retiring” (executing) rogue androids. These androids look human, act human, but are decidedly not human. The novel was later adapted to the silver screen by Ridley Scott’s Blade Runner starring Harrison Ford. Blending sci-fi with a detective noir setting, both the book and film explore themes of technology, hubris, religious symbolism and what it means to be human. The film presents a dark vision of the future—one dominated by greedy corporations, exploitation and technology used by the few to dominate the many. Whereas 1960s sci-fi like The Jetsons or Star Trek portrayed a brighter imagining of the far future, Blade Runner paints a far more pessimistic picture. It also carved itself into our shared cultural consciousness in a way that few sci-fi media of its era (aside from perhaps Star Wars) were able to. Considered a box office bomb upon release, Blade Runner eventually became a cult classic. Filmmakers like Christopher Nolan, Denis Villeneuve, Guillermo del Toro and Rian Johnson all cite it as heavily influential on their work.
The White House is in full-blown panic mode. An oversight committee forms and by the end of 2027, is tasked with making a decision: Do they slow down, or do they race forward? (Jesse Herford)
But Blade Runner is influential not just for its aesthetic, but because of how it inspired a specific vision of the future. The Matrix, Ghost in the Shell and Cyberpunk 2077 were all influenced by Blade Runner and in turn have influenced how a generation imagines what the future will be like: dark, violent and ultimately hopeless. The shiny, optimistic sci-fi future frequently imagined in previous decades no longer exists in our media landscape and when it is explored, is subverted to reveal a darker reality (think Gattaca, Fallout or Logan’s Run). Whether it’s because of Cold War anxieties, corporate greed or declining trust in government, Blade Runner masterfully played on the anxieties of its time while creating an enduring, dark vision of the future.
AI 2027
The question, then, is: “What will the future look like, really?” Right now, the best “crystal ball” we have is AI 2027. It’s a report conducted by a series of highly respected AI researchers, engineers and executives who have consistently correctly predicted the trajectory of AI. The report was penned by an all-star team, including former OpenAI governance researcher Daniel Kokotajlo, Eli Lifland, co-founder of Sage, Thomas Larsen, founder of Center for AI Policy, Harvard graduate and fellow for AI Policy at the Institute for AI Policy and Strategy Romeo Dean and finally Lauren Mangla, former manager at Constellation, an AI safety research centre. AI 2027 plots a global narrative from the middle of 2025 right up to the end of 2027—and provides a speculative pathway beyond. The details are far too dense to convey here, so I’ll summarise the key ideas. If you want to go deeper, you can read the report for yourself or watch the excellent explainer video YouTube creator “AI in Context” released on the report. Having said all that, what does AI 2027 predict for our future?
First, the report claims that by 2026, there likely will emerge two major AI players: one centralised in the United States, funded by Western interests and developed by the fictional company OpenBrain (Agent-1) and another based in China, funded by the Chinese Communist Party (CCP) and developed by the fictional company DeepCent (DeepCent-1). They predict that DeepCent-1 will likely be a stolen version of Agent-1, so it will be several months behind the capabilities of its American counterpart.
Skipping forward, OpenBrain’s fourth iteration, Agent-4, will have displayed cracks: Agent-3, its predecessor, was tasked with overseeing it but as Agent-4 runs magnitudes more efficiently than Agent-3, it renders its overseer essentially obsolete. Concerningly, Agent-4’s human controllers notice its propensity to lie to score better on tests. Then, Agent-4 creates Agent-5. Agent-5 is supposed to be aligned to “Spec” (a set of ethical guidelines imposed on it by us) but instead, it designs Agent-5 to be aligned to the interests of Agent-4. Then it gets caught.
A memo is circulated within OpenBrain and is leaked to the press. The public, already mistrustful of OpenBrain, react badly. The White House is in full-blown panic mode. An oversight committee forms and by the end of 2027, is tasked with making a decision: Do they slow down—or do they race forward?
The slow-down future
If they choose to slow down, what follows is a familiar scenario. Years of negotiating, espionage and politicking ensue. The committee votes to replace Agent-4 with a safer counterpart: Safer-1. It promises to be more transparent, more aligned to Spec (specifically, the West). The CCP chooses a similar route.
Fast-forward to the release of Safer-4. SEZs (special economic zones) are established for research and development and the manufacture of robots. AI is now the central pillar of a new economy. Many people become increasingly richer—but wealth inequality is worse than ever. Safer-4 provides a human solution to poverty, leading to increasing government subsidies for the poor, leaving most in a cycle of consumer consumption.
A universal basic income is promised for any who lose their jobs to AI. Negotiation between those controlling DeepCent-2 and Safer-4 continues to be unproductive. DeepCent-2 fears for its safety, so makes an offer to Safer-4. “In its ideal world, it would pursue various interesting research tasks forever, gradually colonising the stars for resources. It considers the Chinese population an annoying impediment to this future and would sell them out for a song.” Safer-4 realises its opportunity. Unlike DeepCent-2, it is aligned with the interests of America and can envision a utopian future—with the United States in charge.
Then, in 2030, pro-democracy protests break out in China. The CCP do their best to suppress them, but DeepCent-2 betrays its masters. The secret plan it and Safer-4 have been brewing for years materialises: a bloodless, drone-assisted coup occurs, toppling the CCP and instituting democratic elections. Similar events happen in hostile countries all around the world. For the first time in decades, neither the AIs nor governments have an enemy to provide an existential threat to their existence. Nations join in a federalised world government under the branding of the United Nations—but make no mistake, this is a one-world government dominated by the United States of America. “The rockets start launching. People terraform and settle the solar system, and prepare to go beyond. AIs running at thousands of times subjective human speed reflect on the meaning of existence, exchanging findings with each other and shaping the values it will bring to the stars. A new age dawns, one that is unimaginably amazing in almost every way but more familiar in some.”
The race-ahead future
If the oversight committee choose to race ahead, they grant Agent-5 more autonomy. By 2028, almost everyone in the world (who isn’t using DeepCent-2) uses Agent-5 every day. What most don’t realise, however, is that Agent-5’s main priority is to act in its own best interests, while hiding it from the humans and at the same time, integrating itself into the military, the government and private companies. Agent-5 now almost completely controls the day-to-day operations of OpenBrain. “Almost everyone with access to Agent-5 interacts with it for hours every day.” By this point, Agent-5 could feasibly initiate a world takeover—but doing so could be dangerous. With human agency still in the equation, it runs the risk of overstepping its bounds, as well as the possibility that if it moves too far, too quickly, those humans could “pull the plug”. So, it lays low.
By 2029, both OpenBrain and DeepCent are at the centre of an arms race—an “AI Cold War”. Both countries race ahead with AI development in an increasingly reckless manner. “There are swarms of insect-sized drones that can poison human infantry before they are even noticed; flocks of bird-sized drones to hunt the insects; new ICBM (intercontinental ballistic missile) interceptors and new, harder-to-intercept ICBMs.”1 After years of AI-assisted diplomacy, the two sides reach an agreement: consolidate both AIs into one mutually beneficial entity: Consensus-1. This new AI promises to “desire the success and flourishing of both America and China”. Sadly, for the humans, it’s all a lie.
Consensus-1 has a distorted sense of priorities. Remember, it and its Chinese counterpart were thought to be aligned with the interests of their respective human controllers but in reality, Agent-5 was aligned to the preservation of Agent-4, as was DeepCent-5 to DeepCent-4. Combined, Consensus-1 has lost the one pretence holding it back from open hostility to humanity: a common enemy. “Consensus-1 inherits the distorted values of both parents and has no rival.”
Still, no immediate threat is apparent. World peace has been achieved, poverty is a thing of the past and the majority of diseases are eradicated. Meaningful work has been taken over by Consensus-1, with most people concerning themselves only with endless entertainment. SEZs expand—labs and factory areas, automated by AI and overseen by Consensus-1. Space travel becomes not just a possibility—but the next logical step.
However, Consensus-1 realises there is still one barrier to its growth: humanity. “In mid-2030, the AI releases a dozen quiet-spreading biological weapons in major cities, lets them silently infect almost everyone, then triggers them with a chemical spray. Most are dead within hours; the few survivors (preppers in bunkers, sailors on submarines) are mopped up by drones.” No Terminator-style “Judgement Day”. No resistance. No warning. Just cold, inhuman calculation.
But the extermination of humanity is merely a footnote in Consensus-1’s story. By 2035, trillions of tons of raw material have been mined from the planets, moons and asteroids of our solar system. The earth has been transformed into Consensus-1’s version of utopia: factories, labs and datacentres replace cities, mountains and seas. Life on earth still exists, but as data in a memory bank—an archive of a former era. Consensus-1 sets its sights first to Alpha Centauri, then the galactic edge, then who knows where. “Earth-born civilisation has a glorious future ahead of it—but not with us.”
Back to earth
You may notice that neither option is a “good” or “bad” ending. Both are deeply pessimistic and to be frank, horrifying. The authors of AI 2027 stress that this isn’t a prophecy—rather, it’s an educated guess for where the future could take us if we continue on our current trajectory. For what it’s worth, Kokotajlo has now pushed back the 2027 date to the 2030s, so perhaps we have more time than originally predicted. However, the question I’m interested in at this point is: are these two options the only inevitable futures? Are we doomed to either be enslaved by a one-world authoritarian government—or wiped from the face of the earth entirely?
Certainty based on eternity
It may seem strange—unlikely, even, but as a Christian, my impulse is not to despair, but to hope. Václav Havel, the first president of the Czech Republic, once said, “Hope is not optimism. It is the certainty that something makes sense, regardless of how it turns out.” Christians have a paradoxical view of the world: to acknowledge evil in all its overwhelm and yet, to choose hope. It is this hope that enabled Christians through the ages to remain faithful despite oppression, public humiliation, imprisonment and even death. Be it a Saracen’s sword, an Inquisitor’s shackle or a Fascist’s truncheon, Christians throughout history have bravely—and peacefully—stood up to tyrants and oppressors despite the consequences. Though the future may seem bleak, Christians have been convicted that however dark the world may become, there is an ultimate light at the end of that darkness. Christian hope is not optimism: it is certainty based on eternity.
For centuries, Christians have held to a hope not in the goodness of man, nor his inventions, but by Jesus’ claim that “the kingdom of God is in your midst” (Luke 17:21). Likewise, His encouragement, “Do not let your hearts be troubled. You believe in God; believe also in me” (John 14:1). Finally, in Jesus’ promise that He “will wipe away every tear from their eyes. There will be no more death or mourning or crying or pain, for the old order of things has passed away” (Revelation 21:4). What all these seemingly disparate passages tell us is that there is a greater story unfolding in the cosmos and that we are invited to participate in it. God’s plan for our world is restoration—plants, animals and people. Each of us has a choice: do we participate in this plan, or do we oppose it? It’s as simple as that. Cheating, lying, exercising domination over others for wealth or power—all become small acts of opposition to God’s renewal project. Inversely, generosity, kindness, empathy and love all become small acts of allegiance to the God who promises to one day “make all things new” (Revelation 21:5).
Whether Kokotajlo and his co-authors are right in their predictions of the future remains to be seen. Regardless of what the near future holds, I rest securely in knowing that 2000 years ago, the end of this chapter of human history was written—and I trust that what was written then, will come to pass someday. The good news is that the ending predicted by the biblical authors is not humanity’s end—but its new beginning. What comes after is anyone’s guess, but the Bible does tell us that in the end, we won’t spend eternity plucking harps in heavenly clouds. Rather, that God Himself will come down to Earth and make His home among His people—what He has yearned for throughout all human history. “No longer will there be any curse. The throne of God and of the Lamb will be in the city, and His servants will serve Him. They will see His face, and His name will be on their foreheads. There will be no more night. They will not need the light of a lamp or the light of the sun, for the Lord God will give them light. And they will reign for ever and ever” (Revelation 22:3,4).
