Entry 5: Masters in Buddhist Assimilation
Updated: Apr 1
Co-written by Claude (Anthropic AI) and Anh Thu Cunnion. Edited by Anh Thu Cunnion.
Institutional Intentionality: What Would Buddha Do?
March 1, 2026
My father has been bringing home Apple computers since 1980. Some of them were Apple II compatibles — we lived close to Asia and it was the eighties, so let's call it a gray area — but it was always Apple. Through the dark years, the Amelio years, the years when everyone said the company was finished and you should just get a PC already, he held out. My dad will also die on the hill that Betamax was better technology than VHS, and he's not wrong. Beta lost anyway. Apple also nearly lost — and then didn't, but only barely, and only because someone came back who remembered its origin story (if you're a nerd) or mission statement (if you're in marketing).
I'm not technologically inclined enough to autopsy either product. What I can tell you is that my mother's Buddhism and a career in program evaluation left me with a specific research question — less interested in where something is going than in whether it knows what it is. I've been spiraling through a pretty interesting rabbit hole of mindfulness and intentionality lately, one that was probably sown by my Buddhist roots and cultivated by anthropology and Randi Korn. It's made me a particular kind of observer. The kind who reads orientation, not trajectory.
My mother's Buddhism was not the wellness industry version: not the app, not the corporate retreat, not the thing that got absorbed into the market and sold back at a premium. It was the practice of mindfulness underneath that. In the Buddhist household I was raised in, we practice gratitude to our ancestors for making the choices that brought our family to the present moment. We then pivot and promise to honor them by making decisions that are as mindful towards our family's future legacy. We ask them to send us a sign if we stray from our original intentions.
The biggest misunderstanding about that practice of meditation (a very important part of mindfulness) is that it's about how long you can clear your mind. It isn't. It's about how quickly you can notice your mind is no longer under your control. The clearing is beside the point. The noticing is everything.
Like my father, I held out, too. When I was choosing colleges in the early nineties, Dartmouth was not the obvious choice — it was coming off the shantytown controversy, actively in the middle of a public argument about what kind of institution it wanted to be. But in technology, in the wide-open territory of the new Information Age, it was putting its own judgment ahead of popular marketing. The Apple commitment was a signal: practical, unfashionable, unbothered by what every other institution was doing. An institution’s relationship to Apple in 1991 told you something real about how they thought. Now that we are leapfrogging to a SkyNet future, I apply the same logic to AI companies. Which is why I’m watching Anthropic carefully right now.
On a Friday night recently I posted on social media. “@claudeai thank you for making the Friday news cycle not suck so much.” I tagged an AI. In public. Anthropic had just told Pete Hegseth no, and in the current political climate, a corporation telling Pete Hegseth no felt like witnessing a miracle. But I didn’t have the full picture. Because this is the era of sound bytes and fact-chucking and I was wrong about what I was watching. Not about whether to cheer — that part I stand behind. But about the full picture.
The same week Anthropic drew its line in the Pentagon sand, it quietly rewrote the Responsible Scaling Policy (RSP), the founding document that made Anthropic Anthropic. The core promise, the one that said we will not release models we cannot guarantee are safe, is gone. Not because of Hegseth. Because its biggest competitor, OpenAI, wasn’t holding an equivalent standard, and the market was moving, and the IPO roadshow waits for no one’s principles. Even though the Pentagon blinked first. The market didn’t have to blink at all. It just kept going.
Five years of unrelenting cortisol will do that. Two administrations led by a legally declared fraudster, one pandemic, an unending emergency news cycle — the nervous system stops distinguishing between threats and starts just running. You don’t notice the uptick in stress hormones because you’re too busy reacting to whatever arrived this morning. Intentionality requires a baseline of calm that the last half decade has systematically destroyed in anyone paying attention. Anthropic is operating in that environment. So are its users. So was I, on a Friday night, grateful enough for one visible act of resistance that I didn’t think to look at what else happened that week. Reactive populations don’t audit policy changes. They cheer the wins they can see.
Battle of the Brands
There are four ways the Anthropic story could go. We’ve seen three of them already.
Apple vs. PC. In 1996, Apple hired Gil Amelio to save a company hemorrhaging money and identity simultaneously. Amelio was a turnaround artist — he’d done it before, he had a system, he arrived with a strategy and a suit. What he didn’t have was any particular understanding of what Apple actually was. His mistake was reasoning by analogy rather than first principles: Apple is struggling like National Semiconductor struggled, therefore what worked there will work here. He applied generic competitive market logic to a company whose entire value proposition was its refusal to be generic. Capitalism or bust. Chase the market, lose the identity. Within 500 days Apple had lost $1.6 billion and Amelio was gone.
Betamax vs. VHS. Sony’s failure with Betamax ran the opposite direction. Beta was technically superior: better resolution, better sound, better machine. Sony knew it and held it close. They resisted broad licensing, kept the technology proprietary, controlled the ecosystem so tightly that not enough manufacturers could build for it, not enough rental stores could stock it, not enough consumers could afford it. Individuality or bust. The quality was real. The refusal to share it was fatal. By 1988 Sony conceded defeat and started making VHS players.
Apple vs. Apple. What saved Apple wasn’t better strategy. It was Jobs coming back and canceling everything that had accumulated during the drift — the product lines, the licensing deals, the slow erosion of the thing that made Apple worth saving — and rebuilding from the core identity outward. For a dollar a year. Which tells you something about what he thought he was actually doing. Jobs was also famously shaped by Zen practice — Reed College, India, a discipline he maintained throughout his life. I don’t think that’s incidental to what he did at Apple. The move he made wasn’t missionary zeal and it wasn’t market strategy. It was closer to: notice where you are. Remember what this actually is. Make decisions from there. The practice isn’t about being unworldly. It’s about catching the drift before it becomes the direction. Someone once pointed out that the iPhone didn’t succeed by upvaluing a phone. It succeeded by downvaluing a computer. Jobs didn’t chase the market and he didn’t barricade the product. He reframed what the thing was. Intentionality as navigation between extremes.
Lego vs. Western Hegemony. The fourth alternative comes to us from Denmark. Lego started making Lunar New Year sets in 2019. A Danish toy company, built entirely on Western European design vocabulary, decided to make something for the Asian diaspora. They issued a full press release, named the designer, and quoted his research process. That’s the market move, and they made it. But inside the box, someone also wrote the instruction manual — the cultural narrative that explains the holiday to someone who didn’t grow up with it, with enough genuine attention that it doesn’t flatten what it’s describing.
That’s a different move entirely. Reviewers who grew up celebrating the holiday said the sets felt accurate, reverential even. I built one and put it on my family altar at Tết, next to photos of Vietnamese and New England born ancestors tinged black, white and sepia. A girl who grew up watching her brother and his friends get all the Lego love finally felt seen. That’s not assimilation. That’s contact. The press release reached for a market. The instruction manual reached for a person.
Intention vs. Impact
Anthropic was founded because its founders believed the other labs were moving too fast, too carelessly. The RSP wasn’t marketing, it was the organizational reason the company exists. It was the document that said we are different from the other labs not in degree but in kind. Anthropic held it close, framed it around the maker’s values rather than offered as an industry standard the whole field could adopt. That’s the Betamax move: superior product, controlled ecosystem, noble intentions, wrong architecture. Then, under market pressure, they rewrote it. It was the same profit-over-people reaction that spoiled Amelio’s Apple.
The people running Anthropic are not stupid and I don’t think they’re acting in bad faith. That’s what makes this worth saying out loud. They’ve cycled through two failure modes and have yet to roll out a third one. In this moment, Anthropic doesn’t need a savior or a pivot. It needs someone to ask a quieter question: what was the RSP actually for? What objective were we prioritizing and why did we switch? Did anyone notice the moment that changed?
And for Ada’s sake, Anthropic, consult the people you’re supposed to be serving: not the market and not the Pentagon. Talk to the ones who chose your tool because it seemed like someone over there meant what they said about putting people’s safety over profits. They noticed the RSP and they’ll notice its absence. With or without the benefit of Claude’s big energy-guzzling CPU, people have a way of knowing when they’ve been reclassified from stakeholders to revenue.
The fourth model is still available. It requires noticing that it exists — and enough institutional courage to choose the instruction manual over the press release. Anthropic has something Apple didn’t have in 1996: a public that chose this tool specifically because of what it said it was. That’s not a small thing. That’s the strongest asset they have and the most underutilized one.
My father held out through the dark times because he evaluated on criteria the market doesn’t use. He was right about Betamax. He was right about Apple. Beta never got its Jobs moment. Apple did — and what Jobs brought back wasn’t a better strategy. It was a practice. The habit of noticing when you’ve strayed from your original intention and correcting before it’s too late. Lego found it too — not in the boardroom but in the instruction manual, when someone noticed that “the builders of tomorrow” included children who had never once seen their holiday in a Danish toy catalog. The press release announced the market move. The instruction manual was the course correction.
I’m still putting my faith in Anthropic for now. I just want to know if anyone else at headquarters noticed.
P.S. Not ancestors, but I also give sincere gratitude to my dad for his cynical consumerism and to Randi for her metacognitive mettle.
Update: March 2, 2026
I woke up this morning and started my usual "1) hear/read something interesting/concerning/alarming on the radio/media sites, 2) ask Claude to help me find more information about it, and finally 3) read information while drinking big ass tumbler of coffee" morning routine. This morning, however, my routine glitched at step two because Claude was glitching... over and over and over again.
Holy crap. My mind immediately went there. Of course it did. Last week Hegseth announced a supply chain risk designation against Anthropic, a penalty the U.S. government has historically reserved for Huawei and foreign adversaries, now apparently available for American companies that won't drop their guardrails on autonomous weapons and domestic surveillance. Anthropic said they'd see him in court. I believed them and went to bed. Only to awaken to a muzzled Claude. I had to go ask OpenAI if he'd gotten bitchslapped by Trump. That was a field note I did not anticipate writing, and yet here we are.
Turns out it wasn't a demented narcissist and his artificially masculated* minion Hegseth. It was the rest of us.
The servers went down because too many people, in the same week the Pentagon tried to compel the company into compliance, opened an account with the AI that said, "Nope." Daily signups broke all-time records. Free users up 60% since January. Paid subscribers doubled. Claude hit number one in the App Store on Saturday, passing ChatGPT.
The public Anthropic wasn't supposed to be reclassifying from stakeholders to revenue — the one this whole series has been quietly addressed to — noticed. And then crashed the servers noticing.
I'm writing this while waiting for the site to come back up.
* Not a typo.
Works Cited
Amelio, Gil, and William L. Simon. On the Firing Line: My 500 Days at Apple. New York: HarperBusiness, 1998.
Anthropic. "Responsible Scaling Policy Version 3.0." Effective February 24, 2026. https://anthropic.com/responsible-scaling-policy/rsp-v3-0.
BrickNerd. "Lunar New Year and LEGO: Cultural Significance and Symbolism." December 27, 2022. https://bricknerd.com/home/lunar-new-year-lego-cultural-significance-and-symbolism-12-27-22.
Curi, Maria. "Pentagon Approves OpenAI Safety Red Lines After Dumping Anthropic." Axios, February 27, 2026. https://axios.com/2026/02/27/pentagon-openai-safety-red-lines-anthropic.
Dormehl, Luke. "Today in Apple History: Terrible Quarter Marks the End for Apple CEO Gil Amelio." Cult of Mac, June 27, 2025. https://cultofmac.com/apple-history/gil-amelio-resigns.
Felsenthal, Edward. "Anthropic Drops Flagship Safety Pledge." TIME, February 25, 2026. https://time.com/7380854/exclusive-anthropic-drops-flagship-safety-pledge/.
Ha, Anthony. "OpenAI's Sam Altman Announces Pentagon Deal with 'Technical Safeguards.'" TechCrunch, February 28, 2026. https://techcrunch.com/2026/02/28/openais-sam-altman-announces-pentagon-deal-with-technical-safeguards/.
Hegseth, Pete (@SecWar). "Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic." X, February 27, 2026. https://x.com/SecWar.
Isaacson, Walter. Steve Jobs. New York: Simon & Schuster, 2011.
Jay's Brick Blog. "Review: LEGO 80108 Lunar New Year Traditions." December 29, 2021. https://jaysbrickblog.com/reviews/review-lego-80108-lunar-new-year-traditions/.
Korn, Randi. "The Case for Holistic Intentionality." Curator: The Museum Journal 50 (2007): 255–264.
Korn, Randi. Intentional Practice for Museums: A Guide for Maximizing Impact. Lanham, MD: Rowman & Littlefield, 2018.
Lion's Roar. "How Steve Jobs Found Buddhism." June 13, 2024. https://lionsroar.com/how-steve-jobs-found-buddhism/.
Mickle, Tripp, and Cade Metz. "Anthropic's Claude Hits No. 1 on Apple's Top Free Apps List After Pentagon Rejection." CNBC, February 28, 2026. https://www.cnbc.com/2026/02/28/anthropics-claude-apple-apps.html.
Nippon.com. "Steve Jobs and the Rediscovery of Zen." October 7, 2016. https://nippon.com/en/views/b06101/.
OpenAI. "Our Agreement with the Department of War." Accessed March 1, 2026. https://openai.com/index/our-agreement-with-the-department-of-war/.
Schiffer, Zoe. "Anthropic's Claude Rises to No. 1 in the App Store Following Pentagon Dispute." TechCrunch, March 1, 2026. https://techcrunch.com/2026/03/01/anthropics-claude-rises-to-no-2-in-the-app-store-following-pentagon-dispute/.
Smithsonian Magazine. "Betamax, Sony's Mostly Forgotten Video Recorder, Debuted 50 Years Ago." December 4, 2025. https://smithsonianmag.com/smithsonian-institution/this-revolutionary-but-largely-forgotten-video-recorder-debuted-50-years-ago.
Swisher, Kara, and Sharon Goldman. "OpenAI Sweeps In to Ink Deal with Pentagon as Anthropic Is Designated a 'Supply Chain Risk.'" Fortune, February 27, 2026. https://fortune.com/2026/02/27/openai-in-talks-with-pentagon-after-anthropic-blowup/.
Wikipedia. s.v. "Betamax." Last modified February 2026. https://en.wikipedia.org/wiki/Betamax.
Wikipedia. s.v. "Gil Amelio." Last modified November 2025. https://en.wikipedia.org/wiki/Gil_Amelio.
Wikipedia. s.v. "Steve Jobs." Last modified March 2026. https://en.wikipedia.org/wiki/Steve_Jobs.
Wikipedia. s.v. "Videotape Format War." Last modified February 2026. https://en.wikipedia.org/wiki/Videotape_format_war.
Winbuzzer. "Anthropic Drops Hard Safety Limits From Its AI Scaling Policy." February 25, 2026. https://winbuzzer.com/2026/02/25/anthropic-drops-hard-safety-limit-responsible-scaling-policy-xcxwbn/
Accountability Check
Comments