Escape Velocity and the business of the AI future
Dixon, R.J., 2026. Escape Velocity and the business of the AI future. [online] robdixon.ai, 14 October. Article. In brief. Anabology's Escape Velocity sets a fear many business leaders now carry to music: a window is closing, and whoever hasn't moved by the end of it is left behind for good. I don't believe AI creates a permanent underclass. I think it reorders the ladder to success, and that the long-run destination is abundance. But I do think the next eighteen months, as AI moves from adoption into the disruption of markets and supply chains, will largely decide who climbs, who falls, and who gets to shape what comes next.
So, you've probably got AI near the top of your agenda, and you're hearing two stories at once. One promises extraordinary productivity. The other says your business model is about to disappear. Both arrive with the confidence of someone who has already seen the future.
That whiplash runs right through SLOPCORE: Escape Velocity, a music video by anabology. Beneath the fashion imagery and electronic soundtrack sits an argument about work, wealth, and what people will be for when machines can do much more of the doing. You don't need to catch every Silicon Valley reference to feel it.
My reading is that the song stages a contradiction. AI might free us from economic insecurity and even some biological limits. The same technology might deepen exclusion or create dangers we can't control. The music lets us feel both possibilities, then leaves us to live with the uncertainty (anabology, 2026a). I want to take that seriously, because I think it's the right feeling for a leader to have right now, and because I'm going to disagree with part of it.
Technology culture on the catwalk
The verses parade a run of fashion looks, mixing computer spending, corporate ambition, autonomous vehicles, security failures, and longevity. It's a technology industry dressing itself in the future while its products and claims remain fallible. The choruses swing between alarm and renewed confidence.
This is satire from inside the culture, and the confident presentation is part of what we're being asked to examine. You'll recognise the pattern from your own inbox: a supplier announcement, a striking demonstration, a large valuation, and the suggestion that anyone who hesitates has already missed out.
The early wordplay links advanced intelligence with computer-generated imagery and a television detective programme, making futuristic terminology sound like entertainment branding. Artificial general intelligence, usually shortened to AGI, means a proposed AI able to perform a broad range of intellectual tasks at human level or beyond. Superintelligence means capability substantially beyond ours. Both are future possibilities, and the song's enthusiasm for them isn't evidence that today's systems have arrived.
The shifting labels invite the only commercial question that matters: what can this actually do in your business? A more impressive name doesn't answer that.
Eighteen months, and the third wave
The song's recurring countdown imagines a shrinking window in which to escape permanent economic exclusion. It starts at eighteen months, falls to six, and reaches zero.
The song offers no evidence for that number. It's a dramatic device. But I'll be honest: it feels about right to me, and here's why.
For a while now I've described the AI revolution as three waves, each lagging the one before (Dixon AI, 2026a). The first is the technology itself, which keeps compounding. The second is adoption by people and organisations, running six to twelve months behind it. The third is the disruption of markets and supply chains, and until recently it had barely started.
The first wave is in full exponential. OpenAI says its research organisation now gets 3.1 agent-workdays of effort for every human workday (OpenAI, 2026). The second is well underway. In a single month this autumn I delivered sessions and keynotes from Lahore and Dhaka to Chicago, Geneva, and Glasgow, and I didn't meet one business leader who wasn't working out how their business needs to evolve (Dixon AI, 2026b). And the third is starting to show. In September Meta launched Muse, a personal agent that books travel, sends email, and pays at checkout (Meta, 2026). Within a fortnight Amazon had blocked it (GeekWire, 2026). The argument has moved on from whether agents will shop to whose agents, on whose terms.
That third wave is where things get genuinely unpredictable. The first two waves happen inside your business: better tools, new skills, faster work. The third happens to the market around you. Suppliers reprice, intermediaries disappear, customers delegate their buying, and the supply chains you've planned around start to move. That's why I think the next eighteen months matter so much. They're roughly the time it takes for wave three to arrive in earnest, and the period in which the people and organisations who will shape it get selected.
None of this means every job is about to go. The ILO's 2025 assessment found that one in four workers worldwide was in an occupation with some exposure to generative AI, and judged transformation more likely than wholesale replacement (Gmyrek et al., 2025). Those are estimates of exposure, not observed redundancies. For you, the work is to look inside each role, see what's changing, and prepare your people for the responsibilities that remain.
Not an underclass: a reordering of the ladder
The title joins several ambitions. Physical escape velocity is the speed needed to break free of a body's gravitational pull. Economically, the song imagines escaping a position from which you can never catch up. Biologically, it invokes longevity escape velocity, the hypothesis that medical advances might extend remaining life faster than time consumes it. That idea has a serious history, including Aubrey de Grey's discussion in PLOS Biology, but it remains a proposed trajectory, not a promise (de Grey, 2004).
The final chorus changes the economic story. The countdown ends by declaring the underclass abolished. Read alongside the medical imagery, it suggests that sufficiently powerful technology might make the race for security unnecessary. That connects with Dario Amodei's vision of enormous intellectual capability applied to scientific and medical problems, an essay that imagines substantial benefits while recognising the limits set by experiments, hardware, and institutions (Amodei, 2024).
Here's where I part company with the fear in the song. I don't sign up to the idea of a permanent underclass. My view is that we're heading for a reckoning, and a reordering of the ladder to success.
Some very established organisations, and some countries and people, will find that the skills and capabilities that put them near the top now push them back down the stack. Others, who embrace this technology early, will move up. Once that journey starts it'll be very hard to catch up, because the gap compounds at the speed of the technology. What will probably happen is that some organisations die, and their people move to the ones that are rising. New organisations will be created. Early movers will have a huge advantage.
So I'm on the abundance side of this argument. In the long run, I think we end up with abundance for all. But the people and organisations with the agency and influence to shape that world are probably being selected within this eighteen-month window. By the end of it, if you're not one of the disruptors, you're likely to be part of the disrupted. And once you're being disrupted, flipping from one group to the other becomes borderline impossible, because the speed of change leaves no time to rebuild.
That's the real escape velocity for a business. Not breaking free of the economy, but building enough momentum to stay on the rising side of the reordering.
The value of knowing what's worth doing
One idea in the song gives business leaders a particularly useful handle: taste as a moat. A moat is a competitive advantage that rivals find hard to cross. Taste here means the judgement to recognise worthwhile work and choose what deserves to reach a customer.
Think about your proposal team. If AI can produce several plausible drafts in minutes, another draft adds little. Understanding the client, spotting a misleading promise, and choosing an offer you can actually deliver matter enormously. More output raises the value of selection.
That's the argument of my From Knowledge to Purpose paper. As cognitive execution becomes easier to obtain, value moves towards deciding what to pursue and judging the result. Purpose, Execution, and Judgement describe those three activities. It's a conceptual framework whose propositions need testing, not a forecast that every occupation follows the same path (Dixon, 2026d). It also explains why the ladder reorders. Organisations built to reward reliable execution now compete on purpose and judgement, and the capabilities that got them to the top aren't the ones that will keep them there.
The way the video was made illustrates the point. Its creator credits AI tools alongside their own direction and shares a production account. You can judge the finished work, but the account alone can't tell you how much time, intervention, or preparation someone else would need (anabology, 2026b).
So look beyond the demonstration. Can your team repeat the result? Who checks it? What does it cost once revisions and errors are included? An impressive output is a reason to experiment, then measure.
Abundance still leaves a question of ownership
Even on my optimistic reading, the song's hopeful ending leaves a mechanism unexplained. More productive capacity doesn't tell us who receives the income, owns the infrastructure, or gets access to the benefits. Abolishing an underclass takes more than a technical breakthrough.
Sam Altman's Moore's Law for Everything addresses that directly. He argues that AI could shift power from labour towards capital, and proposes wider participation in ownership to share the gains. That's his policy argument, not one the song makes (Altman, 2021). Some go further still. Leopold Aschenbrenner argues that automating AI research could create a reinforcing cycle of ever more capable AI. It helps explain the belief in rapid progress surrounding the song, but it's a scenario with assumptions and bottlenecks, not a law (Aschenbrenner, 2024).
At company level the question is immediate, and it's yours. Will the savings go to customer value, employee opportunity, or margin? Will your people get tools and time to learn, or experience change only as a demand to produce more? Who climbs the ladder inside your business depends partly on choices you make.
Then there's the market. In Who's doing the buying? I distinguish AI helping people work from AI agents doing the work of transactions, and propose the Agent Economy Index to measure the shift (Dixon, 2026j). An agent is software authorised to take actions towards a goal. At the end of September the index put the world reading at around 0.17 per cent, below its 1 per cent early-warning line (Dixon AI, 2026b). That's low, but wave three won't arrive gradually. When your customers delegate their buying, you'll need to compete for their agents' consideration as well as their attention, and sellers who aren't ready won't lose share slowly. Their new-customer pipeline will simply stop.
Your eighteen months
Escape Velocity holds our attention because its exhilaration sits right next to a fear of losing control. I share the exhilaration, and I think the fear is pointed at the wrong outcome. The risk for your business isn't a permanent underclass. It's ending up on the wrong side of the reordering while the window is still open.
So start this month. At your next leadership meeting, pick one valuable workflow. Ask the people who do it to test what AI can contribute, agree how quality will be judged, and decide in advance who should benefit if it works. Then ask the wave-three questions. What share of your revenue is already discovered, shortlisted, or bought through an AI? And could an agent transact with you today? That gives you something more useful than a countdown: evidence to act on while acting still counts.
References
Dixon, R.J., 2026d. From Knowledge to Purpose: The Shifting Bottleneck of Value Creation and Five Domains of Human Value as We Approach AGI. SSRN Working Paper. Available at: https://dx.doi.org/10.2139/ssrn.6396778 [Accessed 14 October 2026]. Companion article: From Knowledge to Purpose.
Dixon, R.J., 2026j. Who's doing the buying? An index for the agent economy. [online] robdixon.ai, 15 September. Available at: https://www.robdixon.ai/post/agent-economy-index [Accessed 14 October 2026].
Dixon AI, 2026a. AI Newsletter August 2026 (A word from Rob). [online] dixonai.com, 23 September. Available at: https://www.dixonai.com/post/ai-newsletter-august-2026 [Accessed 14 October 2026].
Dixon AI, 2026b. Dixon AI Newsletter September 2026 (A word from Rob). [online] dixonai.com, 30 September. Available at: https://www.dixonai.com/post/dixon-ai-newsletter-september-2026 [Accessed 14 October 2026].
Altman, S., 2021. Moore's Law for Everything. [online] 16 March. Available at: https://moores.samaltman.com/ [Accessed 14 October 2026].
Amodei, D., 2024. Machines of Loving Grace. [online] October. Available at: https://darioamodei.com/essay/machines-of-loving-grace [Accessed 14 October 2026].
anabology, 2026a. Escape Velocity. [song and lyrics] Suno. Available at: https://suno.com/song/6fd3d3e2-8bd1-482e-8761-d970d7052f99 [Accessed 14 October 2026]. Music video: SLOPCORE: ESCAPE VELOCITY.
anabology, 2026b. How ESCAPE VELOCITY was made. [online] Creator's production account, reproduced in a public GitHub mirror. Available at: production account [Accessed 14 October 2026].
Aschenbrenner, L., 2024. From AGI to Superintelligence: the Intelligence Explosion. In: Situational Awareness: The Decade Ahead. [online] June. Available at: https://situational-awareness.ai/from-agi-to-superintelligence/ [Accessed 14 October 2026].
de Grey, A.D.N.J., 2004. Escape Velocity: Why the Prospect of Extreme Human Life Extension Matters Now. PLOS Biology, 2(6), e187. Available at: https://journals.plos.org/plosbiology/article?id=10.1371%2Fjournal.pbio.0020187 [Accessed 14 October 2026].
GeekWire, 2026. Amazon blocks Meta's Muse AI assistant in new standoff over agentic shopping. [online] GeekWire, 21 September. Available at: https://www.geekwire.com/2026/amazon-blocks-metas-muse-ai-assistant-in-new-standoff-over-agentic-shopping/ [Accessed 14 October 2026].
Gmyrek, P., Berg, J., Kamiński, K., Konopczyński, F., Ładna, A., Nafradi, B., Rosłaniec, K. and Troszyński, M., 2025. Generative AI and jobs: A 2025 update. ILO Research Brief, May. Available at: https://www.ilo.org/sites/default/files/2025-05/Research%20brief_GenAI%202025%20Update.pdf [Accessed 14 October 2026].
Meta, 2026. Introducing Muse: The World's First Personal AI Agent Built for Everyone. [online] Meta Newsroom, 8 September. Available at: https://about.fb.com/news/2026/09/introducing-muse-personal-ai-agent/ [Accessed 14 October 2026].
OpenAI, 2026. Research acceleration: the view inside OpenAI. [online] OpenAI, 6 September. Available at: https://openai.com/index/research-acceleration-view-inside-openai/ [Accessed 14 October 2026].
Rob Dixon is Founder and CEO of Dixon AI Ltd and lectures at Manchester Metropolitan University. He writes and speaks on AI capability, strategy, and human-AI collaboration.
Comments