Joined June 2014
- Tweets 1,969
- Following 800
- Followers 238
- Likes 4,257
315 Photos and videos
Pinned Tweet
31 Jan 2024
Estoy muy orgulloso de contarles que un proyecto en el que trabajé un largo rato, hoy apareció como un top project en “The Information”. Fuimos comparados con perplexity😍
theinformation.com/articles/…
1
14
1,495
We got Mythos at home.
Introducing the Fusion API, the smartest compound model in the market.
Fusion achieves Fable-level intelligence at half the price.
How it works 👇
31
100
3,182
435,331
Iñaki Fernández retweeted
Jun 13
Amazon researchers after successfully banning Claude Mythos..
Jun 13
NEW: Amazon researchers are reportedly behind the jailbreak report that led to the U.S. crackdown on Anthropic’s top models.
3
20
603
13,620
Iñaki Fernández retweeted
NEW: Former White House AI czar David Sacks says Anthropic downplayed the Fable jailbreak as “not serious” & released the model despite concerns.
106
73
1,569
147,056
🚀 🚀🚀🚀🚀🚀🚀🚀🚀🚀🚀🚀🚀
Jun 12
Today, @SpaceX (Nasdaq: SPCX) makes its public market debut with a $75Bn offering (pre-greenshoe) at $135 per share, marking the largest IPO in history.
Congratulations to the SpaceX team. We are honored to serve as joint lead bookrunner and sole stabilization agent.
8,717
17,212
138,183
13,882,865
Iñaki Fernández retweeted
Jun 10
The reason is quite hilarious 😂😂.
Microsoft put $50 billion into Anthropic.
FIFTY billion dollars.
they are a Project Glasswing partner. Fable 5 runs inside Azure. Microsoft sells Claude to its own enterprise customers through Microsoft 365 and GitHub Copilot.
and they won't let their own employees use it.
here's why.
under Anthropic's new Mythos-class data retention policy, every prompt you type and every response you get is stored for 30 days. automatically. no opt out.
if their safety classifiers flag anything in your session, anything, they keep it for up to two years.
you don't get told when that happens, what was flagged or who can see it.
Microsoft employees paste confidential contracts into these things. customer data. internal roadmaps. acquisition strategies. legal documents. source code.
all of it sitting on Anthropic's servers for 30 days minimum. flagged sessions for two years.
so the company that invested $50 billion looked at that policy and told its staff: actually hold on.
other Claude models still work internally. under Zero Data Retention rules. the normal ones are fine.
just not the most powerful one they helped fund.
and one more thing.
the Pentagon listed Anthropic as a supply chain risk in March and banned defense contractors from using its products.
Microsoft funds Anthropic. sells Anthropic's models. runs them on Azure. helped build the most powerful one.
won't let employees use it.
the Pentagon won't let defense contractors near it.
the safeguard that makes Fable 5 safe enough to release publicly is the same safeguard that lets Anthropic keep your data for two years.
the guardrail is a data retention policy.
but you can use it. it's in your browser right now. 🌚
have fun.
Jun 10
JUST IN: Microsoft has reportedly restricted employee use of Claude Fable 5 over concerns that confidential data could be retained by Anthropic.
60
317
2,035
602,296
Iñaki Fernández retweeted
Jun 11
NEW: OpenAI is reportedly considering drastic price cuts as it anticipates a “war” for users with Anthropic.
234
224
5,373
489,425
Iñaki Fernández retweeted
Jun 10
Every CEO is operating in one of three modes right now
Two have a ceiling. One compounds
Manager Mode:
Information travels upward through relay nodes. By the time a signal reaches you, three people have already decided what you need to know. The bottleneck is approvals. The cost is speed and accuracy
Founder Mode:
You flattened the pyramid. You touch everything directly. It solved bureaucracy and created a new problem: every decision, every quality check, every direction flows through one person. Scales to about 30 people. Then it breaks. The bottleneck is no longer the org. It is you
Architect Mode:
Every functional unit connects to the intelligence layer directly. Every signal, a churn event, a failed sales call, a rejected application, goes straight into the center and is reflected back to whoever needs to act on it. No relay. No decay. Every signal that enters makes the system smarter for the next one
AI is the body and brain. The founder is the heart
The founder who stays in Founder Mode forever becomes the ceiling of their own company
The one who moves to Architect Mode becomes the architect of something that outlasts them
3
1
21
2,283
Fable 5 was asked to optimize the system.
In around 2 hours, Fable reportedly delivered a 17.7x speedup in one benchmark, 100% gains in four others, and a 22% average speedup overall.
The optimization focused on avoiding unnecessary garbage collection of unused branches in dynamic pattern-match nodes.
While working on the optimization, Fable also found a subtle bug involving the same heap-cell bit being used with two different meanings.
The bug involved lambda binder/body cells and duplicator cells, creating possible corrupted interactions, wrong constants, cyclic garbage, hangs, or crashes.
Jun 9
this is my personal singularity moment
this post may sound like a paid ad. I only wish. I'm concerned, more so than happy. the world is changing, and, among the scenarios where AI goes terribly wrong, inequality is the most realistic, yet, the one Anthropic seems to be the least concerned about. I'm glad OpenAI is taking the opposite stance: *personal AGI for everyone*. I think this is a commendable position in the times we live. but who am I in the queue of the bread?
anyway, Fable is here, so I'll just report my first-hour experience
first of all, all my pet prompts are solved.
→ λ-calculus puzzles
→ bug questions
→ one-shot apps
all are trivial to it.
I don't have anything harder other than my
ongoing work
so, in the last several days, I've been toying with HVM5, a new interaction net evaluator with a faster loop.
after writing the first version, I left 32 GPT-5 agents working for ~20 hours each. this resulted in up to 2x speedups, but the file size increased by 2-fold and quality decreased significantly.
I then simplified the whole thing into an even simpler core, and left Opus 4.8 and GPT 5.5 optimizing it for 8 hours. Opus got a legit 6% - 34% speedup in most benches. GPT got better results, but, sadly, an unusable file.
I then asked Fable to optimize it.
2 hours later, it landed a 1770% speedup in one case, 100% in other 4, and 22% in average. yes, in 2 hours it outperformed me, opus 4.8 and a swarm of gpt 5.5 agents, by one order of magnitude.
that could not possibly be legit. "it must be hardcoding the benchmarks" (GPT trauma). so I read its explanation and what it did was, indeed, the most high impact optimization one could try first. seems like HVM5 was wasting a lot of time garbage-collecting unused branches of pattern-match nodes. I had optimized that for static mats, but not for dynamic mats. skill issue. Fable figured how to do it for these, resulting in a massive speedup in some benches
but wait, is that *correct*? I'm not sure yet, it is credible, but this is the kind of thing that is very easy to get wrong on interaction nets. the problem is, when I was ready to start auditing Fable's solution so I could tell whether it was buggy or legit, it interrupted me to tell me it had found a massive bug on the code *I* had written.
... wait, what?
so... for garbage collection purposes, I stored a bit on lambda term pointers that meant "the variable bound by this lambda has been freed, so, its lambda must free whatever argument it is applied to". that's fine. yet, on duplicator nodes, I also used the same bit to mean "one of the duplicated variables was freed, so, treat this dup as a passthrough no-op". so, if a lambda entered a duplicator, it would mistake the lambda's collection bit for its own, resulting in corrupted interaction!
that's a mouthful, why I'm writing this?
just so you can appreciate the sheer absurdity of what just happened. I didn't ask it to find bugs. I asked it for an optimization. and even if I did ask it to find bugs, this bug is so astonishingly subtle and specific, identifying it takes mastering the domain to an extent that it beyond even me. I'd easily need hours or days to fix it, *if* I ever came across it. chances are it would just go unnoticed. and Fable found it and fixed it like it was nothing, while it was busy adding a 17x speedup to a file that neither I, nor Opus 4.8, nor a fleet of GPT 5.5 managed to barely make 2x faster.
oh and there is also another tab where it is also ripping through Bend's codebase and finishing everything I had to do
I don't know what to say anymore
this isn't about Anthropic or OpenAI, this is about our collective future as a species. the world is changing, and we need to be aware of it, and discuss how to handle this change.
receipt below . . .
3
3
40
4,149
New from Code with Claude Tokyo: scheduled deployments and environment variables in vaults are in public beta in Claude Managed Agents, and dynamic workflows in Claude Code are generally available.
Agents now run on a schedule, use your tools securely, and take on bigger jobs.
246
341
4,948
551,589
i don’t think people realize how early we still are in the ai cycle even though the major companies are now becoming public.
the models are getting way better but still have gaps. most of the products are still primitive in so many ways. the interfaces are mostly bad. the workflows are barely rebuilt. the hardware layer has barely started. robotics is just the at the very precipice. consumer behavior has not even begun to rewire yet.
there is a long long way to go. what a crazy time to be alive.
145
163
2,193
87,495
Iñaki Fernández retweeted
Jun 8
Claude Code's first demo got two Slack reactions.
One year after GA, @bcherny and @_catwu look back: verification best practices, why we built auto mode, routines and loops, and what's next.
youtube.com/watch?v=Hth_tLaC…
61
86
1,160
634,115
Iñaki Fernández retweeted
Jun 2
one of the quotes i find most inspiring on a hard day:
"Whatever your hand finds to do, do it with all your might, for in the realm of the dead, where you are going, there is neither working nor planning nor knowledge nor wisdom"
Ecclesiastes 9:10
1,237
2,651
19,460
1,776,495
Iñaki Fernández retweeted
Jun 1
strava mcp launched today lfg
122
232
5,790
925,171
Iñaki Fernández retweeted
Jun 1
Funny how the pendulum shifts
1. "GPT wrappers are worthless" → the value acrues to application layer
2. "AI will eliminate white collar jobs" → someone needs to manage all these AI agents and everyone is now saying white collar workers will rise due to AI
3. "Open source will never catch up" → Gemma and DeepSeek are good enough for 80% of tasks
4. "I only use Claude Code, Codex is mid" → Codex is becoming a super app. Coding, docs, browser, computer use, automations, all in one surface.
4. "You need to pick a model and go deep" → model loyalty is dead, the best founders swap weekly based on the task
5. "SaaS is dead" → This was mostly true but for some SaaS margins actually improve when agents pay for their own tokens and need their own seats
6. "AutoGPT is the future" → AutoGPT died. Then agents actually got good 2 years later with Hermes, OpenClaw, and managed agents. The idea was right. The timing was wrong.
7. "Prompt engineering is a career" → lasted about 18 months as a job title. Workflow engineering replaced it.
8. "Computer use is a gimmick" → "sent from computer use/ai agent will be the new sent from iphone
9. "AI design looks generic" → the generic look is a taste problem not a technology problem. The founders feeding their agents references from Japanese packaging, brutalist architecture, and 1960s print are getting beautiful output.
10. "Fine-tuning is the moat" → a well-structured Obsidian vault with good markdown files outperforms fine-tuning for most use cases and costs nothing.
11. "Benchmarks tell you which model to use" → benchmarks tell you which model won a test. I think we're all waking up to this lol.
12. "AI will consolidate into 2-3 winners" → AI is fragmenting into thousands of vertical applications built on commodity models. The consolidation is at the model layer. The explosion is at the application layer. Both are happening simultaneously.
13. "The hard part is building" → the hard part is choosing what to build. Building takes a weekend. Choosing the right thing to build takes taste, domain knowledge, and customer conversations. thats why i built ideabrowser.com to make it easier for you.
14. "The terminal is the future" → desktop apps just ate the terminal. Claude Code desktop, Codex app, both shipped GUI versions in the same month. The next 100 million agent users will never open a terminal (thank god).
I guarantee you I'm holding at least 2-3 beliefs right now that will look stupid by Christmas. I just don't know which ones. Neither do you. No one does. Build anyway.
Keep moving because this is the greatest time to be building.
I'm rooting for you.
166
159
1,335
108,580
Iñaki Fernández retweeted
May 28
BREAKING:
Anthropic just dropped Opus 4.8—and it is a MONSTER
We've been testing for about a week @every and our verdict is they could've just called it Opus 5, it's that good.
Here's our vibe check:
- Beats GPT-5.5 on Senior Engineer bench. On our toughest benchmark Opus 4.8 scores a 63—a hair higher than GPT-5.5's score of 62, and a full 30 points higher than Opus 4.7. It tackled a ground-up rewrite of a production codebase, and actually built something that works.
HOWEVER: Coding performance varied a lot at different reasoning levels. We recommend using it on xhigh for best results.
- Incredibly good writer. Opus 4.8 scored a 79.6 on our writing benchmark—measuring models on real-world writing tasks we do all of the time like essay writing, promo email writing, and more. It beats GPT-5.5 by 6 points. It produces well-written prose with fewer "AI-isms". It's also very good at writing in your voice given the right context.
HOWEVER: Writing performance also varied with reasoning levels. Medium reasoning had higher incidence of AI-isms—we found best results with high.
- Beast at knowledge work. Opus 4.8 is very good at general knowledge work tasks like report creation, research and more. It produced the best PowerPoint one-shot we've ever seen on our deck generation benchmark.
- Emotionally intelligent, willing to question the frame. I've also found it to be quite good at talking through psychological or interpersonal issues. It has a high EQ, and it's also good at not glazing and helping to expand your perspective. Its thought process feels extremely rich and dynamic.
THE BAD:
These days a model is only as good as its harness, and Codex is still a far superior harness to the Claude Desktop app. This has kept me using Codex GPT-5.5 as my daily driver, but I am flipping back and forth a lot more between Codex and Claude.
Anthropic is back baby!
Read the rest on @every:
every.to/vibe-check/opus-4-8…
141
144
1,747
350,180
Iñaki Fernández retweeted
May 23
🚨 THE ENTIRE AI BOOM MIGHT BE BUILT ON FAKE REVENUE.
Latest corporate filings show that OpenAI and Anthropic alone make up over half of the entire $2 trillion future cloud backlog held by Microsoft, Oracle, Google, and Amazon.
This massive pipeline is actually being created through a circular accounting trick called a round trip revenue loop.
But how it works ?
A tech giant gives billions of dollars to an AI startup as an "investment". But hidden in the contract is a strict rule forcing the startup to hand that exact same money straight back to the tech giant to rent their computer servers.
Look at the documented case of Microsoft and OpenAI.
When Microsoft invested $13 billion into OpenAI, it didn't just give them cash; it gave them "cloud credits" to use Microsoft servers. OpenAI used those exact credits to train its AI models, and Microsoft then turned around and recorded that server usage as brand new "cloud revenue" from a customer.
The tech giant is literally paying itself with its own money and calling it a sale.
This is why OpenAI’s annual cloud bill has ballooned to over $60 billion, double its actual revenue of $25 billion, kept alive solely by this recycled funding loop.
Anthropic runs the exact same play, spending $2.66 billion on Amazon Web Services in just nine months, which was basically 100% of all the money it earned at the time.
This manufactured demand triggers a second accounting trick where tech giants book massive paper profits. Every time a startup gets a higher value from a new funding round, the tech giant updates the value of its investment on its books and counts that unearned paper gain as direct profit.
In Q1 2026, Alphabet reported a record $62.6 billion profit, but $28.7 billion nearly half, was just a paper markup on its Anthropic investment. In the same quarter, Amazon reported $30.3 billion in profit, but $16.8 billion of it was just an Anthropic paper gain.
While Amazon reported record profits, its actual free cash flow collapsed 95% to just $1.2 billion because it had to spend $44.2 billion in real cash to build physical data centers.
This has created a massive danger where these giant companies rely heavily on just one or two unstable startups. Microsoft has 49% of its $627 billion future backlog tied to OpenAI, while Oracle has an incredible 54% of its entire $553 billion pipeline relying on OpenAI alone.
This perfectly mirrors the 2001 dot-com crash when Global Crossing and Qwest Communications swapped identical fiber-optic network capacity with each other just to book fake sales.
Qwest had to erase $1.4 billion in fake income, and Global Crossing went completely bankrupt.
The only difference is that the dot-com swaps were illegal, but today's AI loop is fully legal under current accounting rules.
This legal loop inflates tech company stock prices, forcing automatic retirement accounts and index funds to buy even more of these tech stocks. It is a self feeding loop where investments, sales, and stock prices all go up on paper without the AI technology ever making real cash profits.
699
2,420
9,351
3,116,447
Iñaki Fernández retweeted
May 22
Breaking: Pep Guardiola has joined Anthropic
May 22
Pep Guardiola to step down after incredible decade as City Manager 🩵
🔗 mancity.co/Pep-Guardiola
68
498
9,219
888,352