A Supercomputer Pointed At Your Brain: Mindfulness, Machines & Metrics

Every platform you use points to the same argument: you clicked, so you must have wanted it. Randima Fernando thinks that's the biggest lie the attention economy tells about itself — the same reflex that snaps a driver's eyes toward a car crash on the highway, mistaken for desire. Build a feed that manufactures car crashes all day and you will get clicks. You will not get anything anyone actually wanted.

Randima Fernando was raised by Buddhist parents, spent seven years building a nonprofit that scaled mindfulness into schools, and has spent eight more advising Silicon Valley on humane design. So it's a strange thing to hear him tell Thomas Hübl, flatly, that meditation alone is not going to save you. Not because mindfulness doesn't work — because the thing on the other side of your screen is, in his words, a supercomputer pointed at your brain, and no amount of discipline reliably beats an opponent with more data on you than you have on yourself. That's not where he leaves it, though: his actual answer runs on two tracks at once, one aimed at what a person can do inside their own life, and one aimed at what has to change in the system itself.

In This Summary

Part One

Monetizing Fundamental Life Units

"Attention," Randima says, "is the fundamental unit the mind operates in" — and an entire industry exists because that unit turned out to be divisible, and sellable. Free products exist to harvest the slices, then resell them to advertisers, which means a platform isn't optimized for what you actually came to do. It's optimized for how long it can hold you before handing you off to someone else.

That single misalignment cascades outward. When TikTok found a faster way to hijack attention with short-form video, every competitor was sent, in his words, "back to the boardrooms and the strategy sessions" to copy it or lose ground — a race that rewards whatever captures attention fastest, with no mechanism to ask what it costs the person on the other end. Journalism drifts toward the "click-baity, hyperbolic, divisive" end of the spectrum because that's what survives the algorithm. Political campaigns run A/B tests and learn, empirically, that the most outrage-generating headline wins — so they run more of them. A "like" button quietly teaches a platform what to show you more of, and a feed narrows, year over year, into an increasingly extreme version of whatever already resonated. His suggested experiment is almost unsettlingly simple: swap feeds with a friend who votes differently than you do. You'll see, he says, "very different content and very different reality." Children pick up the underlying "social physics" of likes and shares the way they learn gravity — early, and by feel, not instruction. Randima draws the same extraction logic out to the edge of the planet: growth built on "billions of air conditioners, billions of cars, billions of livestock" behaves the same way attention extraction does — both eventually outrun what the underlying system, a mind or an atmosphere, can absorb.

Part Two

Reflex, Not Desire

Thomas brings his own frame to the pattern: a regulated nervous system, he suggests, makes for a more nuanced relationship with technology — more capacity to notice what's off, more real choice. A dysregulated one, trauma tipping someone into hyperactivation, numbness, or fragmentation, narrows that same field before a person even gets to choose. He asks Randima directly: is the attention economy learning to exploit that internal dysregulation on purpose?

Randima's answer runs through addiction as a matter of depth — how far, in his words, "the jack" goes into the brain. These technologies have learned to trigger the dopamine responses that evolved to reward survival and reproduction, and aim them at nothing: a notification, a scroll, a flash of light. The clearest case, for him, is children — not because anything is wrong with them, but because their dopamine systems are still calibrating to the world. Put a screen in front of a two-year-old and their eyes will involuntarily track it, because the brightness has been engineered to exceed everything else in the room — what he calls a "hypernormal stimulus," beyond anything a body evolved to expect. Once a child's dopamine receptors have calibrated to that level of input, ordinary life reads as flat, and taking the device away can trigger real, disproportionate distress.

He's careful about a distinction most conversations about technology skip past: what people click is not the same as what they want. "People clicked on it, so that's what they want," companies say — but that, Randima argues, confuses reflex with intention.

Everyone looks at a car crash on the highway, he says, because attention to danger is a survival mechanism, not because anyone actually wants a world full of car crashes. His close is structural, not moralizing: a business that survives on your attention has a built-in conflict of interest with actually serving you, because a genuinely useful product would give you what you needed and let you leave — the opposite of what keeps the metrics up.

Part Three

Limits of Extraction and Shifting the System

Asked how the system actually changes, Randima splits the question cleanly in two: what's possible inside the current economy, and whether that economy is sustainable at all. Inside it, he says, everything eventually runs through price — so the interventions that work are the ones that eventually touch a company's bottom line. A documentary like The Social Dilemma builds public awareness; awareness becomes political pressure for new laws; laws enable litigation; litigation finally attaches a dollar figure to harms that had been free to inflict. Whistleblowers matter for the same reason — a leaked internal document is proof a company knew, which is what turns awareness into liability. The obstacle, always, is speed: technology moves faster than any of these correcting mechanisms can track it.

Zoom out further, and Randima applies the same law to the mind that he'd apply to oil.

Take oil out of the ground faster than millions of years can replace it, and that's a problem. Slice attention, focus, and sense-making faster than rest and reflection can rebuild them, and — by the same logic — that's a problem too. Hard-won capacities like sustained attention take real discipline to build and very little to erode; a weekly meditation practice, he notes, is no match for a phone engineered to compete for attention eight hours a day or more. Children, whose minds are more malleable, feel it fastest.

He connects the thread to inequality with a line that doubles as a diagnosis: capitalism "is very sensitive to capital" — genuinely responsible for real advances in medicine, food, and technology, and just as reliably concentrating wealth unless something actively counteracts it. He sees the arrival of powerful AI less as a new problem than as the old one at higher speed — and, possibly, as the pressure that finally forces the incentive structure to change.

Part Four

AI: Extraction Problem, Now Faster

Randima's frame for artificial intelligence is almost deceptively simple: technology has always been humanity's greatest accelerator, and this generation of AI is the most powerful version of that accelerator yet — compounding gains across hardware, algorithms, data, and compute all at once, faster than Moore's Law ever moved. There's an irony he doesn't dwell on: the same graphics hardware he spent seven years at Nvidia building, for making pictures on a screen, "lighting up pixels," turned out, almost by accident, to become the engine of the very acceleration he's now warning about.

But an accelerator doesn't change direction. It amplifies whichever direction a system is already pointed in. Pointed at a predominantly extractive economy, AI doesn't redirect the extraction — it puts it "on overdrive." The nearest effect, he argues, is automation of cognitive work before physical labor, for the plain reason that software is easier to automate than robotics. The business logic isn't sinister so much as mundane: machines don't need sick days, vacations, or healthcare, so wherever a company can cut cost by simplifying, it will — the same structural gravity that made trickle-down economics fail to actually distribute anything.

He's careful not to make this a story with only one ending. AI's upside — in medicine, in climate solutions, in new sources of energy — is real, he says, and shouldn't be waved away. His caveat is about sequence: society has to make it through a doorway of good sense-making and wise collective choices before those benefits can actually land, and if the underlying systems break down first — polarization, inequality, a shared reality already fractured — the good outcomes never arrive at all.

What makes this second contact more dangerous, in his account, is distribution: this technology isn't centralized the way social media platforms were, which means the regulatory tools that struggled even with a handful of large companies are up against something running on everyone's phone — anyone can now generate a convincing fake video, or a fake research paper, at will.

Part Five

Win It by Not Being in the Fight

Closing out, Randima offers two distinct moves — not sequential, not either/or, but different enough that each deserves its own weight. One is aimed at the system. One is aimed at the self.

Shift the Incentives

To understand what any system, company, or person is actually optimizing for, read the incentive structure, not the stated words: incentives reveal much more about behavior than words, or advertising campaigns. It's the same bottom-line logic covered above in Part Three — price mechanisms, legislation, litigation, whistleblowers turning quiet awareness into real liability. Changing what a company gets rewarded for doing is slower than changing a habit, but it's the lever that touches the system itself, not just one person's relationship to it.

Renunciation

The personal move starts before any of the rest of this: get genuinely clear, ideally out loud, with people and teachers you trust, about what a wise life and real thriving actually mean for you — and let that clarity, not the algorithmic menus of Netflix or social media, set the direction of your attention.

Thomas turns the conversation back to mindfulness directly: what role does it actually play here? Randima's answer has two parts, and this is where the claim from the top of this piece gets its full accounting. The first part is generous: mindfulness builds real awareness — of thought, emotion, bodily sensation — which creates the internal space to notice a choice before it's made instead of after. Paired with prior clarity about what a skillful life looks like, that pause is where better choices actually happen.

The second part is the harder one, and it's the same one he opened with: even advanced mindfulness training is not going to do that much for you if you're standing inside a system built by a company running constant analysis against fundamental human cognitive bias — a supercomputer pointed at the brain, as he puts it. His closing counsel isn't to out-discipline the algorithm. It's a kind of renunciation, a deliberate separation from whatever's already been identified as unskillful, rather than a battle to be won by trying harder.

Not winning. Declining to fight at all — and using whatever clarity mindfulness provides to see, more plainly, where not to stand.

Neither move stands in for the other. Shifting incentives changes what the system rewards; renunciation changes your own relationship to that system while it's still what it is. Randima has spent years now doing both at once.

↑ Back to top