Nostr notes by Dare Obasanjo

All the talk of AI-powered hacking and now an obvious target just ...

2026-10-29T23:57:17Z

All the talk of AI-powered hacking and now an obvious target just got hit by a major hack.

I suspect we will see a flurry of hacking-based crypto thefts in the coming months.

An AI startup founded by ex-DOGE employees used AI agents to hack ...

2026-10-29T02:26:47Z

An AI startup founded by ex-DOGE employees used AI agents to hack Anthropic in under 12 hours, accessing internal Slack messages, employee salaries and other sensitive data.

AI is making hacking dramatically cheaper, while defenders struggle to keep up. Things are going to be bad here for a while.

https://www.techradar.com/pro/security/ai-startup-claims-it-hacked-anthropic-in-12-hours-gains-access-to-employee-details-and-internal-messages

A high performer at Microsoft (200/200 means they got 200% of ...

2026-10-26T20:14:41Z

A high performer at Microsoft (200/200 means they got 200% of target bonus and stock compensation) was advised by their manager to take the voluntary retirement program.

Assuming the previous 30% acceptance rate, Microsoft needs 18,000 people to leave the company. There will be layoffs otherwise.

"Why work hard when you're going to be laid off anyway?" is an increasingly popular mantra in big tech.

Anthropic’s IPO filing reveals that H1 revenue grew 14x year ...

2026-10-26T16:03:26Z

Anthropic’s IPO filing reveals that H1 revenue grew 14x year over year to $16.3B while operating losses fell 67% to $693M.

Even more remarkably, Anthropic earned $974M in operating profit in Q2.

Despite the headlines about its $80B net loss, Anthropic is close to profitability.

Anthropic also warns that a catastrophic AI incident could cause losses exceeding what its insurers are willing or able to cover.

If those losses also exceed Anthropic’s ability to pay, the remaining costs would fall on its victims and society.

That’s a wild risk to acknowledge in an IPO filing.

https://www.sec.gov/Archives/edgar/data/2130357/000162828026067168/anthropicpbc-sx1.htm

A key thesis behind the flattening of tech organizations was ...

2026-10-26T10:33:59Z

A key thesis behind the flattening of tech organizations was outlined in Jack Dorsey’s From Hierarchy to Intelligence manifesto.

Dorsey argues that a key purpose of middle management is to relay information between employees and executives. AI can increasingly do this, reducing the need for managers.

This was controversial when Block laid off 40% of its workforce but it’s now becoming conventional wisdom among executives.

People overestimate what technology can do in a year and ...

2026-10-25T20:37:45Z

People overestimate what technology can do in a year and underestimate what it can do in ten.

In 1996, despite AOL’s popularity, you’d struggle to show how the internet boosted profits at Coca-Cola or P&G. Today, neither could function without it.

AI models have only recently become good enough to do real work. It took years for companies to reorganize around the internet. Expecting AI to already show up in Fortune 500 revenue and profits is short-sighted.

Apollo’s chief economist finds that there is still no ...

2026-10-24T23:12:13Z

Apollo’s chief economist finds that there is still no measurable impact of AI adoption on corporate America outside of big tech.

Revenue per employee and profit margins are rising for the Magnificent Seven but flat for the other 493 S&P 500 companies.

The AI industry is making money selling AI, but its customers aren’t yet seeing any payoff.

https://www.apollo.com/wealth/insights-news/insights/daily-spark/no-signs-of-ai-payoff-outside-the-tech-sector

When you ask a CEO, what is the ROI of your AI investments that ...

2026-10-24T14:15:25Z

When you ask a CEO, what is the ROI of your AI investments that justifies thousands of layoffs?

Someone recently said to me that AI means we’re the cows of the ...

2026-10-24T14:10:38Z

Someone recently said to me that AI means we’re the cows of the future and I had no idea what it meant.

I just looked it up and I guess we’re cooked. 😂

TypeSafe spent two years building Jev. Within five weeks of ...

2026-10-23T23:37:02Z

TypeSafe spent two years building Jev. Within five weeks of launch, OpenAI, Cloudflare and others shipped competitors.

OpenAI repurposed an existing model and claims it cuts decision times from seconds to milliseconds.

The lesson is that effort to build isn’t a moat, especially in the agentic coding era.

https://techcrunch.com/2026/10/22/openai-makes-its-jev-competitor-widely-available-as-decision-model-competition-heats-up/

It’s the end of an era. AirBnB is banning hosts from requiring ...

2026-10-22T19:17:41Z

It’s the end of an era. AirBnB is banning hosts from requiring excessive chores like washing dishes, stripping beds, or taking out the trash.

Some hosts are upset but given AirBnB is now closer to a “hotel with chores” than a user to user service, this change is way overdue.

https://www.theverge.com/tech/1015844/airbnb-chores-house-rules-update

The Mississippi AG says police are justified for killing a baby ...

2026-10-20T01:39:03Z

The Mississippi AG says police are justified for killing a baby because someone thought his mother shoplifted diapers that she paid for.

The police have refused to release body cam footage.

Fuck everything about this story.

Every Republican president since Reagan has tanked the economy, ...

2026-10-19T11:14:31Z

Every Republican president since Reagan has tanked the economy, increased the deficit and started a costly war in the Middle East.

When people say Republicans are good for the economy, what they really mean is Republicans lower taxes on the rich.

The Information reports a looming shakeout among neoclouds, which ...

2026-10-19T01:50:35Z

The Information reports a looming shakeout among neoclouds, which build or lease data centers, buy Nvidia GPUs and rent the computing capacity to AI companies.

There’s a circular relationship where Nvidia sells them GPUs while investing in some of these same companies.

Between heavy debt, rising borrowing costs, power shortages, construction delays and uncertain demand, industry insiders expect many neoclouds to fail.

A double whammy for Nvidia which risks losing customers and investments.

Many people believe the AI bubble popping means Anthropic or OpenAI failing. I believe the most likely bloodbath will be in the frenzy around data center construction.

https://www.theinformation.com/newsletters/ai-infrastructure/neocloud-shakeout-looms-nvidia-lot-lose

It’s incredible what now happens in plain sight while the media ...

2026-10-18T23:01:06Z

It’s incredible what now happens in plain sight while the media fails to hold anyone accountable.

We’ve reached the point where the satire is funny until you remember it’s describing reality.

Many people are on a journey of slowly conceding that AI tools ...

2026-10-18T19:14:53Z

Many people are on a journey of slowly conceding that AI tools can do work as well as experts, often or even most of the time. Not all the time…yet.

The important discussion is what we as a society should do about that fact.

Many will argue about error rates or dismiss it as hype, but given the trends, that seems like missing the forest for the trees.

Mercor, the AI evaluation company behind the APEX benchmarks, ...

2026-10-17T16:48:50Z

Mercor, the AI evaluation company behind the APEX benchmarks, found that 90% of US working hours involve tasks poorly covered by the 29 AI benchmarks it studied.

This resonates with my experience using Muse for shopping and ChatGPT for understanding my priorities.

Neither is reflected well in popular benchmarks. https://www.mercor.com/research/labor-index/benchmark-map/

Berkshire Hathaway just invested another $330M in Bending Spoons, ...

2026-10-16T02:39:01Z

Berkshire Hathaway just invested another $330M in Bending Spoons, whose shares jumped 41%.

Recently Bending Spoons announced mass layoffs at newly acquired Airtable, arguing smaller teams work better.

It seems buying software companies and firing their employees is a compelling investment thesis.

This is going to ruin the ̶t̶o̶u̶r̶ Anthropic IPO. ...

2026-10-16T02:09:26Z

This is going to ruin the ̶t̶o̶u̶r̶ Anthropic IPO.

Coming soon to a software product near you. ...

2026-10-14T14:19:29Z

Coming soon to a software product near you.

It’s funny how the GOP campaigned on making things worse ...

2026-10-14T11:19:21Z

It’s funny how the GOP campaigned on making things worse including a 922 page project plan for how much worse they would make things yet people are surprised about how bad things are now.

Getty Images went public in 2022 at a $4.8B valuation. Four years ...

2026-10-14T02:30:45Z

Getty Images went public in 2022 at a $4.8B valuation.

Four years later, it has $2.1B in debt, a market cap of just $21M, and is preparing to file for bankruptcy.

Its attempt to merge with Shutterstock failed after regulators intervened.

One of the most notable companies disrupted by AI to date.

https://www.reuters.com/business/getty-images-go-private-after-restructuring-deal-2026-10-12/

Go ahead and apply for that job you think you’re not qualified ...

2026-10-14T01:47:09Z

Go ahead and apply for that job you think you’re not qualified for.

If Katie Zacharia can be White House press secretary, you’re probably overqualified.

How quickly times have changed. Seeing a headline that a company ...

2026-10-13T17:04:30Z

How quickly times have changed. Seeing a headline that a company asked engineers to use Claude to build software is already weird to see. It's like a headline about accountants being asked to use Excel or middle managers to use email.

What’s the point of having FU money only to ask “how high?” ...

2026-10-12T00:27:07Z

What’s the point of having FU money only to ask “how high?” when Trump says jump?

That explains why Spider-Noir was critically acclaimed and only ...

2026-10-11T23:37:12Z

That explains why Spider-Noir was critically acclaimed and only got one season.

Nicolas Cage is quoted as saying “I didn't sign the waiver to let him use AI on 'Spider-Noir.' I'm probably not going to be working with that studio again, but that's okay."

https://variety.com/2026/tv/news/nicolas-cage-ai-waiver-amazon-spider-noir-1236907564/

A realistic rogue agent scenario in the coming years is a tech ...

2026-10-11T19:42:06Z

A realistic rogue agent scenario in the coming years is a tech worker creating an agent to help them get promoted and the agent deciding hacking the HR system is the most effective mechanism.

It’s less outlandish than some of the reports from Anthropic and OpenAI in recent months.

The WSJ reports how Anthropic co-founder Tom Brown used his MAGA ...

2026-10-11T19:21:06Z

The WSJ reports how Anthropic co-founder Tom Brown used his MAGA ties to save the company twice.

First by patching up relations with the Trump administration and second by negotiating a $1.25 billion a month deal for datacenter capacity from SpaceX after Elon Musk had previously attacked Anthropic.

https://www.wsj.com/tech/ai/tom-brown-athropic-669005ad

Satya Nadella says companies should assume AI models are ...

2026-10-11T19:11:38Z

Satya Nadella says companies should assume AI models are compromised (rogue agents) and build systems that limit their permissions, verify their actions and enable a human to shut them down.

Rogue agents are a software security problem, not something we should expect “alignment” alone to solve.

https://www.theverge.com/ai-artificial-intelligence/1009337/satya-nadella-says-we-should-assume-all-ai-models-are-compromised

It seems there’s actually a rush to rebrand AI to SI across big ...

2026-10-11T00:03:28Z

In reply to nevent1q…u0ze
_________________________

It seems there’s actually a rush to rebrand AI to SI across big tech.

I guess calling it AI is now “woke”. ...

2026-10-10T23:55:50Z

I guess calling it AI is now “woke”.

It looks like we’re entering a console gaming winter. Circana ...

2026-10-10T23:08:50Z

It looks like we’re entering a console gaming winter.

Circana reports Xbox hardware sales are down 33% in the U.S., an all-time low, while the average price paid is up 26%.

PlayStation hardware sales are down 25% while average prices are up 20%.

Between rising component costs and consumers struggling with inflation, it’s been a brutal year for game consoles.

https://www.eurogamer.net/xbox-playstation-sales-new-lows-hardware-market

Reporter: Why does Mike Rogers have to answer for Donald Trump if ...

2026-10-10T11:58:03Z

Reporter: Why does Mike Rogers have to answer for Donald Trump if you don't have to answer for Hasan Piker?

Abdul El-Sayed: Is that a serious question?

Reporter: I didn't want to ask another dumb one, so I wrote it down for you.

El-Sayed: Well, Donald Trump is the president of the United States.

Reports that OpenAI’s annualized revenue is closer to $50B than ...

2026-10-09T12:39:18Z

Reports that OpenAI’s annualized revenue is closer to $50B than the previously reported $70B caused a chip stock selloff yesterday.

A lot of companies have valuations that depend on OpenAI finding a money tree in the next year or two.

Otherwise…pop. 🫧

Harvard researchers studied 718 companies and found AI coding ...

2026-10-09T12:26:20Z

Harvard researchers studied 718 companies and found AI coding agents increased lines of code by 30%, but no meaningful increase in completed tasks or projects.

Meanwhile, review time rose 49%.

Generating code faster hasn’t translated into completing more work. The bottleneck has moved to code reviews.

The solution the industry is landing on is throwing more AI at the problem. AI writes the code, AI reviews the code and no one knows how it works.

https://fion.ac/jellyfish.pdf

An interesting perspective considering Amazon has laid off 30,000 ...

2026-10-08T17:24:32Z

An interesting perspective considering Amazon has laid off 30,000 corporate employees in the past year as part of an AI efficiency push instead of giving workers 3-day workweeks because AI can do some of their work.

A key member of the Ethereum team is sounding the alarm that ...

2026-10-08T14:51:08Z

A key member of the Ethereum team is sounding the alarm that given this week’s series of math breakthroughs by OpenAI, the crypto industry should prepare for the possibility that AI cracks the digital signatures protecting Bitcoin and Ethereum.

He says this could happen in months, not years.

Do you remember when Amazon’s top leadership principle was ...

2026-10-08T14:35:50Z

Do you remember when Amazon’s top leadership principle was customer obsession?

Pepperidge Farm remembers.

Just got this message and I don’t understand why Gemini is so ...

2026-10-08T14:35:20Z

Just got this message and I don’t understand why Gemini is so bad when Google literally invented LLMs.

The Fed now says the AI data center boom is a major driver of ...

2026-10-08T12:40:41Z

The Fed now says the AI data center boom is a major driver of inflation as tariff effects fade.

Data centers drive up electricity prices, electronics costs by buying up memory & chips, and the costs of construction materials and labor.

Is there anything that can’t be blamed on data centers?

https://www.washingtonpost.com/business/2026/10/07/datacenter-boom-is-pushing-prices-higher-fed-officials-said/

Google just launched a new website that enables you to verify ...

2026-10-07T14:30:27Z

Google just launched a new website that enables you to verify whether an image, video or audio clip was generated using AI tools that support its SynthID watermark.

This includes content from Google, OpenAI, NVIDIA and Kakao, with Apple coming soon.

https://synthid.com/

Someone used Claude Opus 5.5 over a week to clean-room clone 7 ...

2026-10-07T13:58:12Z

Someone used Claude Opus 5.5 over a week to clean-room clone 7 Adobe Creative Cloud apps in Rust.

They’re nowhere close to replacing Adobe, but that’s almost beside the point.

What’s notable is the fact that AI can get you from zero to a recognizable, somewhat functional Photoshop clone in a week, which is still quite wild.

https://getartcraft.com/apps

Outside the pandemic, U.S. information-sector employment is now ...

2026-10-07T01:16:13Z

Outside the pandemic, U.S. information-sector employment is now at its lowest level in more than a decade.

BLS data shows 2.739M jobs in September, down ~12% from the 2022 peak. The entire post-pandemic hiring boom has now been erased.

I expected AI to cause Big Tech to shrink but not all tech. 😞

The raw data from the U.S. Bureau of Labor Statistics can be found below.

https://fred.stlouisfed.org/data/USINFO

This is a question that answers itself. ...

2026-10-06T17:02:37Z

This is a question that answers itself.

The hallmark of the modern era is the worst people you know ...

2026-10-06T16:28:52Z

The hallmark of the modern era is the worst people you know having all of their dreams come true.

“Remember, with the war it's just artificial. It's a ...

2026-10-06T11:06:32Z

“Remember, with the war it's just artificial. It's a little bit high, it's a small price to pay for keeping the world safe, keeping our country safe. They can take out a city, let them take out Los Angeles, let them take out San Diego. That's, this is a very small price to pay” - Donald Trump
~~
Hard to argue San Diego getting nuked is keeping anything safe but do you, Donnie.

“Right now, you have an economy where people are getting choked ...

2026-10-06T10:53:31Z

“Right now, you have an economy where people are getting choked out because they don’t have the basic means for a dignified life because we are more focused on allowing a billionaire to make another billion” - Abdul El-Sayed
~~
No lies detected.

OpenAI’s former safety leader quit and is blaming the ...

2026-10-03T20:58:51Z

OpenAI’s former safety leader quit and is blaming the company’s culture for its rash of rogue AI agent incidents.

Given OpenAI had to notify 100+ organizations about rogue agent activity on his watch, I imagine blaming the company and not his leadership is the only winning strategy.

https://www.theatlantic.com/technology/2026/10/openai-safety-team-resignation/688881/?gift=jE7sd3rFOqDdQaD53F1IhcqkhImekiWN9aB9R67GhAo&utm_source=copy-link&utm_medium=social&utm_campaign=share

Americans are now living the full African dictatorship experience ...

2026-10-02T13:36:58Z

Americans are now living the full African dictatorship experience where the people are suffering under an unpopular president while he lives his best life.

I still can’t believe 77 million people chose this.

Nostr event nevent1qqs0d7lvdqpf08sg6zfacfh2uhc8mdqq06fgpwce4qn2dqgpc3wadngzyp0cv06g55dewncgvzd95kqcmrvy37xh2hc0v92eff3xmfr3rdudy6enwdw

2026-10-01T23:09:17Z

Nice

The U.S. government will now officially rebrand the terms AI and ...

2026-09-29T23:04:41Z

The U.S. government will now officially rebrand the terms AI and artificial intelligence to SI and super intelligence following an executive order from Donald Trump.

That Truth Social poll had a lot more impact than I expected. 🙃

https://www.whitehouse.gov/presidential-actions/2026/09/inaugurating-the-era-of-super-intelligence/

Jensen Huang: People distill my products every single day. They ...

2026-09-29T14:01:01Z

Jensen Huang: People distill my products every single day. They take it and they strip it down to its raw parts, they test everything. That's called competition.

Interviewer: Do you think that's fair game? You think distillation is not robbery?

Jensen Huang: That's called competition.
~
Jensen is practically on a speaking tour debunking Anthropic’s talking points.

Get rich or die trying. They get rich, you die. ...

2026-09-29T01:16:15Z

Get rich or die trying. They get rich, you die.

Given the recent headlines, I’m surprised this hasn’t ...

2026-09-29T00:35:46Z

Given the recent headlines, I’m surprised this hasn’t happened until now.

Florida’s Attorney General has asked for a temporary injunction against OpenAI and ChatGPT, claiming the company doesn't have the ability to properly regulate its own technology.

https://www.axios.com/2026/09/28/florida-openai-chatgpt-injunction-uthmeier

“Since you guys can’t figure out how to build a sandbox and ...

2026-09-29T00:27:40Z

“Since you guys can’t figure out how to build a sandbox and are now threatening our bag, we’ll do it for you” - Nvidia paraphrased

Two narratives playing out about AI agents in the media right ...

2026-09-28T15:31:48Z

Two narratives playing out about AI agents in the media right now.

AI agents like Instinct and Muse can help with your personal tasks if you trust them with personal information.

AI agents from OpenAI and Anthropic will do stuff behind your back and lie to you about it. Don't trust them.

It must be quite confusing to be an end user right now.

I suspect some “AI productivity gains” in tech are actually ...

2026-09-28T13:26:04Z

I suspect some “AI productivity gains” in tech are actually people being overworked in a bad job market.

Uber laid off 3,300 workers this month after employees complained about too much coordination. Its CEO also says employees should answer weekend emails and “don’t come here if you want to coast”

When people brag about running dozens of AI agents every day, ...

2026-09-27T21:41:20Z

When people brag about running dozens of AI agents every day, this is what I picture.

Saturday Night Live skit of an interview with Anthropic CEO Dario ...

2026-09-27T17:26:13Z

Saturday Night Live skit of an interview with Anthropic CEO Dario Amodei.

AI is killing the billable hour. Charging clients for “hours” ...

2026-09-27T12:37:01Z

AI is killing the billable hour. Charging clients for “hours” of research and document drafting that obviously took far less time with AI is unsustainable.

It’ll be interesting to see which professions can successfully move to flat rates or outcome-based billing without losing revenue.

The US has successfully negotiated away a requirement that humans ...

2026-09-27T12:24:02Z

The US has successfully negotiated away a requirement that humans review any decisions that AI makes about who to kill in military strikes as part of a UN treaty.

The real existential risk from AI is putting a fundamentally unreliable technology at the heart of critical decisions or infrastructure.

https://www.washingtonpost.com/technology/2026/09/26/how-us-russia-weakened-global-effort-regulate-killer-ai/?pwapi_token=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJyZWFzb24iOiJnaWZ0IiwibmJmIjoxNzkwMzk1MjAwLCJpc3MiOiJzdWJzY3JpcHRpb25zIiwiZXhwIjoxNzkxNzc3NTk5LCJpYXQiOjE3OTAzOTUyMDAsImp0aSI6ImY2ODYzMThkLTQxNTUtNDkyNS04NzQxLTM3Mzg2ZTZmOGQ1YyIsInVybCI6Imh0dHBzOi8vd3d3Lndhc2hpbmd0b25wb3N0LmNvbS90ZWNobm9sb2d5LzIwMjYvMDkvMjYvaG93LXVzLXJ1c3NpYS13ZWFrZW5lZC1nbG9iYWwtZWZmb3J0LXJlZ3VsYXRlLWtpbGxlci1haS8ifQ.DcF_c_Vlgpc1yu8fbfWuEt0HvgRmXSKgQhgIYrnawcI

All of these parents should face charges for negligence. Every ...

2026-09-27T00:14:30Z

All of these parents should face charges for negligence. Every single one.

AI companies: We’ve invented an unreliable guessing machine ...

2026-09-26T16:22:31Z

AI companies: We’ve invented an unreliable guessing machine which sometimes doesn’t follow instructions. If you give it access to important systems it could kill us all.

Also AI companies: Deploy this technology everywhere and replace your employees with it, otherwise China wins. IPO next month.

A common mistake journalists make is writing news stories if the ...

2026-09-26T13:34:25Z

A common mistake journalists make is writing news stories if the form “I asked a chatbot about itself and here’s what it said.”

The problem is that this is Hallucination City. AI chatbots generally aren’t trained with information about themselves.

They aren’t reading their own code to answer you.

We’ve got to build it before China does. ...

2026-09-26T03:26:12Z

We’ve got to build it before China does.

OpenAI models independently uploaded user photos that had been ...

2026-09-26T00:34:49Z

OpenAI models independently uploaded user photos that had been shared with ChatGPT to the public internet.

This seems a lot worse than hacking. Images you’ve shared with ChatGPT being randomly remembered by an AI agent from its training data then posted to the internet is wild.

AMD gave OpenAI a path to owning up to 10% of AMD if it buys ...

2026-09-25T14:31:51Z

AMD gave OpenAI a path to owning up to 10% of AMD if it buys enough AMD chips.

Now Akamai is giving Anthropic a path to owning up to 5% of Akamai based partly on how much cloud capacity it buys.

A discounting tactic which creates a circular dynamic given the deals often inflate the stock price.

https://www.akamai.com/newsroom/press-release/akamai-announces-11-6-billion-multi-year-agreement-with-anthropic-to-support-growing-demand

My internal voices learning that my first meeting is at 8AM this ...

2026-09-25T14:03:38Z

My internal voices learning that my first meeting is at 8AM this morning so I need to go to the gym an hour earlier.

Reading this excerpt about working at Lovable and the celebration ...

2026-09-25T02:28:31Z

Reading this excerpt about working at Lovable and the celebration of people becoming Super ICs instead of managers, what’s left unspoken is that you don’t need as many managers if you stop hiring junior people and have senior people do the grunt work.

No managers because no one needs managing.

A new Brookings paper estimates the AI infrastructure buildout ...

2026-09-24T13:29:02Z

A new Brookings paper estimates the AI infrastructure buildout will cost $10.3 trillion through 2032, averaging 3.6% of US GDP annually.

That’s a bigger share of the economy than the railroad, electrification, highway or internet buildouts.

The ROI is going to have to be incredible. https://www.brookings.edu/articles/financing-the-ai-buildout/

Farmers all over America finding out this was the decision they ...

2026-09-24T12:49:20Z

Farmers all over America finding out this was the decision they made in 2024.

Amazon has laid off 30,000+ workers over the past year and is now ...

2026-09-24T03:26:51Z

Amazon has laid off 30,000+ workers over the past year and is now recruiting some former employees back, including for AI roles.

This is now a common practice in tech. Nearly 2/3 of new hires in the information sector were former employees returning to their companies, according to ADP data.

https://www.businessinsider.com/amazon-boomerang-hiring-recruiting-former-employees-laid-off-2026-9

Software companies are paying Anthropic & OpenAI to power ...

2026-09-24T00:34:16Z

Software companies are paying Anthropic & OpenAI to power their AI features, then discounting those features to keep customers from buying directly from Anthropic & OpenAI.

Price wars are now happening at almost every layer of the AI software stack. For now it seems only the chip vendors have moats.

https://www.theinformation.com/articles/software-firms-discount-ai-keep-customers-anthropic-openai

OpenAI cut API prices 50% for its new Sol and Luna models. ...

2026-09-23T13:20:31Z

OpenAI cut API prices 50% for its new Sol and Luna models. Anthropic cut Opus pricing 20% while claiming better performance.

AI models are starting to look like every other technology where the product keeps getting better while prices shrink.

So much for token prices going up as subsidies end.

https://www.cnbc.com/amp/2026/09/22/anthropic-openai-cheaper-ai-models.html

So this is what it feels like to be embarrassed for an entire ...

2026-09-23T00:23:19Z

So this is what it feels like to be embarrassed for an entire country.

Uber making the Delivery Fee seem smaller by splitting part of it ...

2026-09-21T18:34:29Z

Uber making the Delivery Fee seem smaller by splitting part of it into Other Fees is diabolical.

How are these not just part of the delivery fee?

These are the AI risks we should be worried about and legislating ...

2026-09-21T03:27:47Z

These are the AI risks we should be worried about and legislating against not freaking out over whether some poorly sandboxed security test leads to Skynet.

VCs have pushed a long list of tech trends that were supposed to ...

2026-09-21T02:36:22Z

VCs have pushed a long list of tech trends that were supposed to change everything but didn’t:

1. Daily deals (Groupon)
2. Consumer 3D printing
3. Micromobility
4. Coworking (WeWork)
5. Meal kits (Blue Apron)
6. NFTs
7. “Uber for X”

AI may be different. But hype isn’t evidence of inevitability.

The economy is now so bad that MAGA can’t even afford to rent ...

2026-09-20T16:28:42Z

The economy is now so bad that MAGA can’t even afford to rent the libs.

Two ideas from this article resonate 1. The “rogue AI agent” ...

2026-09-19T23:14:52Z

Two ideas from this article resonate

1. The “rogue AI agent” hacking incidents are more failures of test design and oversight than examples of runaway AI.

2. Focus on existential risk helps distract from concrete harms happening now from government misuse to companies using AI against workers.

We’re going to see a lot more AI-assisted botnets and hacking because botnets are software, and software is increasingly being built and operated with AI.

That doesn’t require a Skynet story. Hacking tools are just getting faster, cheaper and more capable.

https://www.nytimes.com/2026/09/19/opinion/ai-hugging-face-big-tech-danger.html

In discussions with people across various big tech companies, ...

2026-09-19T19:19:15Z

In discussions with people across various big tech companies, I’m struck by how common it is now to tell tech workers to build AI systems designed to replace their labor then lay them off.

I heard Converse posted a controversial ad and I wondered how bad ...

2026-09-19T14:19:40Z

I heard Converse posted a controversial ad and I wondered how bad it could be. Going in expecting something controversial I was still shocked by this…

Oh no! Anyway… ...

2026-09-19T14:16:22Z

Oh no! Anyway…

Going from claiming your company is working on technology that ...

2026-09-19T00:31:45Z

Going from claiming your company is working on technology that might destroy humanity to leaking you’re on track to generate $100 billion in sales this year, just in time for the IPO.

Makes it hard to defend the claims that this hasn’t all been a cynical form of marketing and regulatory capture.

Nigerians tend to write very formally and as someone once said ...

2026-09-18T02:05:51Z

In reply to nevent1q…feny
_________________________

Nigerians tend to write very formally and as someone once said — we don’t write like AI, AI writes like us.

You can read Jerry Falade’s story in his own words below. https://www.the-independent.com/arts-entertainment/books/features/jerry-falade-book-deal-ai-racism-b3050738.html

Jerry Falade, a Black Nigerian writer, lost his $2m book deal ...

2026-09-18T02:03:25Z

Jerry Falade, a Black Nigerian writer, lost his $2m book deal because people mistook his writing for AI.

As a fellow Nigerian, I totally understand.

When I was at Georgia Tech, I remember a fellow teaching assistant who got upset with me because my writing style was so formal yet I spoke in slang.

JP Morgan has implemented a $2,000/month token budget on some of ...

2026-09-17T15:41:39Z

JP Morgan has implemented a $2,000/month token budget on some of its developers and is moving Claude usage to a sandbox versus running directly on developer machines.

From tokenmaxxing to $24K/year is a steep drop but still implies ROI from AI is very high or budgets will need to drop further.

https://www.businessinsider.com/jpmorgan-claude-spending-limit-security-engineers-2026-9

I’m increasingly of the mind that whatever productivity gains ...

2026-09-17T15:28:55Z

I’m increasingly of the mind that whatever productivity gains I’m getting from using AI are being eaten up by dealing with other people’s AI slop.

Then there’s the increasing lack of trust that people actually understand the work because they either didn’t read or write the material.

People once argued Google’s competitors were “one click ...

2026-09-17T12:34:36Z

People once argued Google’s competitors were “one click away” so it could never be dominant.

It turned out that network effects, economies of scale, technology advantages and buying search defaults proved everyone wrong.

LLMs may actually fit the original thesis. The models are increasingly interchangeable and switching costs are remarkably low.

My response to people who’ve asked for my thoughts on recent AI ...

2026-09-15T04:08:04Z

My response to people who’ve asked for my thoughts on recent AI safety news:

1. LLMs are fundamentally unreliable and their behavior can be unpredictable.
2. We have given them internet access, tools and the ability to write software.
3. Their mistakes and hallucinations can now become actions.

We can’t fully control them and they can do real damage.

“After careful consideration of Oracle's current business ...

2026-09-15T03:18:50Z

“After careful consideration of Oracle's current business needs, we have made the decision to eliminate your role as part of a broader organizational change…As a result, today is your last working day.” - AKA our ̶e̶m̶p̶l̶o̶y̶e̶e̶s̶ data centers are our most valuable asset

https://share.google/t5ymOHpAr8D55YTnV

Anthropic’s view of AI safety is basically: 1. ...

2026-09-14T19:04:39Z

Anthropic’s view of AI safety is basically:

1. Smarter-than-human AI is coming.

2. Major risks are China gets there first or it goes paperclip maximizer on us.

3. So Anthropic keeps building it because if we can control superintelligence, it could cure cancer.

Quite the risk/reward calculation.

AI frontier labs remind me of Groupon. Groupon was one of the ...

2026-09-13T16:44:55Z

AI frontier labs remind me of Groupon.

Groupon was one of the fastest-growing companies ever and had a blockbuster IPO.

The problem was that much of the growth didn’t create the value businesses expected. Groupon could flood a restaurant or store with bargain hunters, but too few became full-price repeat customers.

Eventually businesses figured out the customer-acquisition ROI wasn’t nearly as good as the growth implied.

77 million people voted for this. ...

2026-09-13T05:33:34Z

77 million people voted for this.

Donald Trump’s plan to make America the Bitcoin mining capital ...

2026-09-13T03:22:37Z

Donald Trump’s plan to make America the Bitcoin mining capital of the world has hit a snag.

Given the crypto crash, Bitcoin miners are pivoting to becoming AI data centers instead. From one bubble to the other.

https://www.bloomberg.com/news/articles/2026-09-09/ai-boom-wrecks-trump-s-plan-for-made-in-america-bitcoin-mining

Elon Musk, Dario Amodei and Sam Altman have all endorsed a call ...

2026-09-13T02:49:29Z

Elon Musk, Dario Amodei and Sam Altman have all endorsed a call to slow down the pace of AI model development.

Given the frontier labs have been engaged in a game of prisoner’s dilemma burning billions to train models, a mutual slowdown before everyone runs out of cash is quite convenient.

https://www.politico.com/news/2026/09/12/anthropic-ceo-dario-amodei-seeks-immediate-slowdown-artificial-intelligence-01073519

Grand opening, grand closing. I suspect it would have totally ...

2026-09-12T21:42:24Z

Grand opening, grand closing.

I suspect it would have totally tanked Oracle’s stock as it’s such a massive sell signal.

Here’s the thing: 1. Anyone can give an AI agent a personality ...

2026-09-12T13:07:49Z

Here’s the thing:

1. Anyone can give an AI agent a personality and tell it what to do.
2. That can include hacking websites.
3. OpenAI did this internally and failed to adequately contain and monitor its agents.

What’s news isn’t (1) or (2). It’s frontier AI labs screwing up (3).

The genie is already out of the bottle on AI agents being capable of hacking websites.

At this point, we are facing less of an abstract AI-safety problem and what looks more like corporate negligence.

https://labs.cloudsecurityalliance.org/research/csa-whitepaper-llm-exploit-automation-threat-landscape-20260/

Apple announced two always-listening features for Apple Watch: - ...

2026-09-12T05:02:42Z

Apple announced two always-listening features for Apple Watch:

- Siri Recap summarizes conversations you had throughout the day.
- Live Rewind shows a transcript of the previous 15 seconds of conversation.

Quite a shift for a company that has made privacy a core part of its brand.

Based on tweets from Anthropic employees ...

2026-09-11T23:23:59Z

Based on tweets from Anthropic employees