AI Coding

Built a free IELTS practice tool and looking for testers 🙌 Currently has Reading, Writing, and a CEFR level checker (A1–C2). Would love your honest feedback!
The Main Interesting Part, AI will follow your level, generate your level practices, and while practicing, the coach chat will help you to start, to understand, with ideas, and so on.
👉 https://ielts-studio-five.vercel.app/

Claude Opus 4.8, Sonnet 4.6, models are not working right now! Are we waiting for new models?

Opus 4.8 is back. 😅

Opus 4.8 is not working either?
Bro, what is going on at Anthropic right now?
Their best public model just went down this morning. Elevated errors across the board. Status page says "investigating," — which basically means they don't know what's wrong yet either.
And this is happening in the SAME week where:

Fable 5 got banned by the US government in 3 days
Dario Amodei flew to Washington for emergency meetings
NVIDIA and Google security experts wrote open letters defending them

Now Opus 4.8 is down too.
If you need Claude right now, switch to Sonnet. It's somehow the last thing still standing over there.
I don't know whether to feel bad for Anthropic or just be amazed that one company is having this much happen to them in a single week.
Their engineers woke up this Monday knowing their CEO is in a government meeting fighting to keep their work alive — and their servers decided today was a great day to join the chaos.
Some weeks just don't go your way 😭
Is anyone else getting errors right now? Drop a comment 👇
either

Anthropic's CEO Dario Amodei is flying to Washington, D.C. today to sit across the table from the Commerce Secretary who just killed his product. No press release. No prepared statement. Just two sides of a very tense standoff, in the same room.
Here's the simple version of what happened 👇
Anthropic built Fable 5 — the most powerful AI they've ever released to the public. Three days after launch, the US government banned it worldwide. Not just for China or Russia. For literally every non-American on earth, including Anthropic's own foreign employees.
The reason? Someone reportedly jailbroke it — found a way to bypass its safety guardrails. The government told Dario: fix it or pull it. He said no. So they banned it.
Now here's what most people aren't talking about.
Executives from Nvidia, Zoom, Mercedes-Benz, former US government security staff, and former Google security staff have all signed an open letter demanding the government lift the ban. Their argument is simple and honestly pretty alarming — "pulling the best capabilities away from defenders while our adversaries keep advancing is dangerous."
In other words, by banning this AI to protect us, you may actually be making us less safe.
And the government has only shared verbal evidence of the jailbreak. No documentation. No technical proof. Just a phone call and a 90-minute deadline.
Today's meeting is supposed to finally include actual documentation of the alleged vulnerability. Whether that leads to Fable 5 coming back online — nobody knows yet.
What we do know is this: the most powerful AI ever built is sitting offline right now while a CEO and a Cabinet Secretary argue about whether it's safe enough to exist.
That's the world we're in now. 👇, D.C.

Anthropic's CEO Dario Amodei is flying to Washington DC today to sit across the table from the Commerce Secretary who just killed his product. No press release. No prepared statement. Just two sides of a very tense standoff, in the same room.
Here's the simple version of what happened 👇
Anthropic built Fable 5 — the most powerful AI they've ever released to the public. Three days after launch, the US government banned it worldwide. Not just for China or Russia. For literally every non-American on earth, including Anthropic's own foreign employees.
The reason? Someone reportedly jailbroke it — found a way to bypass its safety guardrails. The government told Dario: fix it or pull it. He said no. So they banned it.
Now here's what most people aren't talking about.
Executives from Nvidia, Zoom, Mercedes-Benz, former US government security staff, and former Google security staff have all signed an open letter demanding the government lift the ban. Their argument is simple and honestly pretty alarming — "pulling the best capabilities away from defenders while our adversaries keep advancing is dangerous."
In other words, by banning this AI to protect us, you may actually be making us less safe.
And the government has only shared verbal evidence of the jailbreak. No documentation. No technical proof. Just a phone call and a 90-minute deadline.
Today's meeting is supposed to finally include actual documentation of the alleged vulnerability. Whether that leads to Fable 5 coming back online — nobody knows yet.
What we do know is this: the most powerful AI ever built is sitting offline right now while a CEO and a Cabinet Secretary argue about whether it's safe enough to exist.
That's the world we're in now. 👇

So what really caused the most powerful AI model in history to be blocked in 3 days flat?

Let me break it down, because this story has more twists than most people are catching.
On June 9th, Anthropic launched Fable 5 — their most advanced AI ever released to the general public. It was a massive deal. Months of hype, thousands of hours of safety testing, and a model so capable it was being compared to having a world-class researcher sitting at your desk 24/7.
Three days later, it was gone. For everyone. Everywhere on earth.
Here's the full story of what actually happened 👇
The Setup
Fable 5 wasn't just another chatbot update. It was built on top of Mythos — Anthropic's most restricted, most powerful model that was previously only available to a tiny group of trusted government and cybersecurity partners. Fable was essentially Mythos with safety guardrails bolted on, making it safe enough — they thought — for the public.
The problem? Someone figured out how to remove those guardrails.
The Jailbreak
A trusted partner — someone who worked closely with both Anthropic AND the US government — discovered they could bypass Fable 5's safety systems. Not completely. Not universally. But enough to potentially unlock the kind of cyber capabilities that Mythos was built for and kept locked away from the public.
They took that finding straight to the White House.
The Ultimatum Nobody Talks About
Here's the part of this story that most news headlines are burying.
The government didn't immediately ban Fable 5. According to David Sacks — the White House AI adviser — they went to Anthropic CEO Dario Amodei first and gave him a clear choice: fix the jailbreak, or pull the model down yourself.
Dario said no.
His argument? The jailbreak was narrow, non-universal, and every AI model in existence — including OpenAI's — has the same vulnerability. Fixing it or pulling the model would set an impossible standard that would essentially freeze the entire AI industry.
That's actually a reasonable technical argument. But the government didn't see it that way.
The Amazon Plot Twist
Now here's where this stops being a tech story and becomes something else entirely.
The trusted partner who discovered the jailbreak and reported it to the White House? According to sources close to the situation, it was Amazon. Andy Jassy — Amazon's CEO — was apparently personally in contact with administration officials about it.
Let that land for a second. Amazon is one of Anthropic's largest investors. They have pumped billions of dollars into this company. And they went to the government behind Anthropic's back.
Whether that was the right call or not, that is a stunning move inside what was supposed to be a close business relationship.
The Ban
Friday evening, 5:21pm ET. The Commerce Department sent a letter to Anthropic. They had roughly 90 minutes to comply. No detailed written explanation of the national security threat. No transparency about what exactly was found. Just an order: shut it down, now, for every foreign national on earth.
Not just in China or Russia. In the UK, Canada, India, France — everywhere. Including Anthropic's own foreign national employees who had been building the model for years.
Anthropic complied. But they didn't stay quiet.
Anthropic Fires Back
In their public statement, Anthropic pushed back hard. They said the jailbreak was narrow and that shutting down a model used by hundreds of millions of people over a non-universal vulnerability makes no sense. They argued that if this standard were applied across the entire industry, it would halt every new AI model release from every lab going forward.
They also made clear this didn't follow their own principles of what responsible government oversight should look like — transparent, fair, and grounded in technical facts.
Translation: we think this was political, not purely about safety.
Why This Actually Matters Beyond the Drama
Here's the bigger picture that nobody is really talking about yet.
This is the first time in US history that the government has issued an export control directive for an AI model itself. Not for the chips that run it. Not for the data that trained it. For the model. Treated like a weapons system. Like a missile or a nuclear technology.
That precedent is now permanently set.
The next time a frontier AI lab releases a model with a vulnerability — and there will be a next time, because perfect jailbreak resistance doesn't exist — the government now has a playbook. They can ban it. Instantly. Globally. With 90 minutes' notice and no written explanation.
So who was right here?
A government that gave a company 90 minutes to comply, provided no written technical evidence, and may have been influenced by a rival investor?
Or a company that spent months safety testing, launched in good faith, and pushed back on a standard they say would cripple the entire industry?
Honestly, both sides have a point. And that's what makes this so uncomfortable.
What I do know is this: we just entered a new era of AI regulation. Not through legislation. Not through a vote in Congress. Through one letter, on a Friday evening, that took down the world's most powerful AI model in under two hours.
Whatever you think about who was right in this fight, that moment changed things.
What happens next time? 👇