The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: https://t.co/bwn0sximKZ
Tons of talk about the gross margin of tokens for models providers, inference clouds, neoclouds, etc.
An equally important topic is the gross margin of tokens for the END enterprise customers. Meaning what's the incremental revenue generated from those tokens.
On the (Continued) Last Art of Face-to-Face Customer Meetings:
@jasonlk has written extensively about this but it's incredible how this still hasn't come back yet. It's the easiest move in B2B. Go see your customers and prospects.
I spoke to a startup founder recently (who's great). He asked about category creation and accelerating sales cycles. I said:
"How often does your team meet your clients face-to-face?"
He said "every now and then if they ask."
Early in Gainsight, when we just launched, we were a tiny company competing with other tiny companies. There was a publicly-traded software firm that was looking at tech for Customer Success and we heard they didn't even consider us. I sleuthed to figure out the decision maker and found one SUPER weak LinkedIn shared connection (my former EA).
I cold emailed the person referring to our common connection and told him I was going to be in his area later in the week and would love to meet. He said "sure - come by."
He was in Charleston, SC. I was in Palo Alto, CA. I immediately Googled how to get to the (lovely) city in SC and bought redeye tickets. Rolled off the plane, freshened up in the airport and met the prospect. Flew home. We ended up winning the deal and amazingly the company is a Gainsight client 13 years later, having spent collectively several million dollars.
My guess is no other vendor "happened to be in the area." That was the only reason we won.
Over the years, your product becomes the best, your brand is the leader, you have trust, etc.
In the early days, all you have is your willingness to get on a plane.
10 Things I Learned at Emergence Capital's AI Native Services (AINS) Summit:
In the world of enterprise AI adoption, the last mile is driving change. One approach is to sell agentic software to companies and to help them change (eg via FDEs). Another is to BE the service provider and use agents to win.
My friend @jakesaper put on an excellent (totally packed) event on this topic. My learnings:
1. Faster/Better/Cheaper: I like Jake's definition - AINS is about becoming a service provider that uses agents to win in one of those three dimensions versus tradition firms.
2. Sell Outcomes: In software, the customer owns the outcome. In AINS, the vendor does.
3. Gross Margins Have to Eventually Be Software-Like: This is a big unknown but is the bet.
4. Get Better with Models: When something like Fable 5 comes out, software companies get scared about their value being eroded - AINS firms get excited about the new leverage.
5. Mirage PMF Is A Risk: One challenge is an AINS firm throws people at the problem, grows fast but then never gets the leverage through agents.
6. McKinsey+Stripe: It's hard to be both a services and software firm - you need "builders" and "doers."
7. Frenemies: Many firms start by delivering part of a service and working through legacy service providers - with the goal of owning the whole process.
8. Users = Employees: In AINS, the software is used by the firm's employees. So the loop of improvement is faster.
9. Repeat Founders: Sometimes it gets boring to go after the same software category over and over again. But now, a repeat founder who built a software business can go after a TAM 10X bigger (due to labor capture).
10. Are They Winner Take All: My biggest question - services industries (eg legal, accounting) are often heavily fragmented. As such, they don't produce breakout, venture-scale winners in most cases. Will AINS make this different and in which spaces?
Awesome event!
Pretty interesting how Anthropic is protecting usage of Claude Fable 5 (at least in Claude Code):
You have to use /model to get to it
It switches back to Opus 4.8 on the next turn (versus staying on Fable in the session)
Token budgets are the hottest topic in enterprise AI right now. And with Claude Fable 5 coming out today, token spending has the potential to 🚀
That might sound fun for developers but it's not as fun for CFOs and companies trying to plan.
So excited for @rfradin and team at @Larridin to launch the first tool for Token Spend & Insights. I'm on the Board along with Alastair (Alex) Rampell and have watched Russ and his colleagues navigate to this existential problem for AI.
The biggest problem holding back AI right now: cost is exploding, and almost none of it is tied to business outcomes.
We all know SOMETHING great is happening. Nobody's quite sure what.
Today we fix that — introducing Larridin Token Spend & Insights.
https://t.co/zgIygTz3ZH
BREAKING:
Anthropic just dropped Claude Fable 5—this is Mythos, made safe for public release. It is the best coding model in the world.
We've been testing it internally @every for the last week or so across coding, writing, marketing, editing, and more—here's our vibe check:
- It broke our benchmarks. Fable scored a 91/100 on our Senior Engineer benchmark—this is human senior engineer level. The previous high score was Opus 4.8 at 63. GPT-5.5 is a 62.
- It's a one-shot wonder. You can set it and forget for hours or overnight on huge coding tasks, and come back to completed work. It cleared entire production bug backlogs, built a playable 3D, and even made a 2-minute animated film—all one-shot.
- Taste and attention to detail. In coding and knowledge work tasks, it has much better taste and attention to detail than we've ever seen. It gets subtle things right, adds little features you might not have thought of, and generally understands the assignment in ways that surprised us.
- Great use of context. We set it loose analyzing customer feedback surveys and our website data and it came back with a crisp, clean report that identified a. our biggest problem and b. a concrete testable solution—and then we sent it off to build that.
- It's best for power users. If you're already used to orchestrating multiple agents in your work, this model can do things that you've never seen before. If you're a knowledge worker or vibe coder with a more basic setup, you're not going to notice a huge difference—in fact, it probably isn't the right model for you.
- It's very slow, token-hungry. Using this thing for regular knowledge work is like squashing an ant with a rocket launcher. It also routinely uses 500k to 1M tokens on tasks. That's why it's best for your heaviest jobs—but not as good for tasks like collaborative writing.
- It's expensive. It's about twice as expensive as Opus, and it's also incredibly token hungry—so expect it to be something you'll use sparingly unless your company pays for it.
Overall, I think of it like a warp drive for coding: It can get you across the galaxy in a few hours, when it used to take months or years. But it's not appropriate for getting around town—you need something faster, cheaper, and more maneuverable.
The ceiling is extraordinarily high on this model though. Even our most advanced testers like @kieranklaassen felt like they were only scratching the surface of it.
Want our full vibe check with all of our testing and benchmarks? Read it on @every: https://t.co/MgJLZszJUB
Say hello to Fable 5, a Mythos-class model we've made safe for general use 🔥
This one's a doozy. So glad we can finally share with the world the ridiculous capabilities we've been relying on so heavily inside Ant the last few months.
Got a cold call from an SDR today (looked like a number of a friend so I picked up).
"Hi Nick - my name is [] - you don't know me and we haven't spoken before."
I listened to the rest of the pitch because the opening was so unique.
Honesty can be disarming in a good way.