Jennifer Martinez (@jenmartinez) header
Jennifer Martinez (@jenmartinez) profile picture

Ex-DC journo who returned West. Formerly @thehill @politico @paloaltontwks @meta @robinhoodapp Now: comms @anthropicai

CA
Joined January 2008
Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)Photo by Jennifer Martinez (@jenmartinez)
Jennifer Martinez retweeted
We’re working with Cursor on this incident, but it wasn’t a compromise of Anthropic’s systems or employee laptops. A malicious actor used access to Cursor’s marketplace to publish a tampered copy of the Claude Code extension, which we didn’t build or publish. The versions on our official channels are not affected and this only impacts Cursor users.
We identified a security incident affecting a modified Claude Code VS Code extension accessed through our Cursor IDE extension marketplace, impacting at most 5,000 users. This malicious version of the extension was not developed by Anthropic. It does not impact other extensions downloaded from the cursor.com/marketplace, Grok Bot, Grok or our other SpaceXSI products. In response, we took down our extension marketplace this afternoon, worked with relevant providers to neutralize the malware, and have reinstated the marketplace after confirming there is no further risk. If affected, please review instructions in this blog post to understand how to find the malicious package and establish a path to mitigate any potential impact. We are asking all customers update to Claude Code for VS Code version 2.1.314: cursor.com/blog/claude-code-…
62
51
15
831
141,811
The mystery has been solved.
Here's a video explaining it:
167
Cool new office - thanks for having us @MTSlive @theojaffee @logangraham
1
132
This very cool - curious to hear what folks think
Replying to @AnthropicAI
To secure open-source software, we’re launching OSS Scanner. We’ll use our frontier models to periodically scan opted-in open-source projects for vulnerabilities, at no cost. Our reports will provide a proof-of-concept, explanation, and suggested fix. anthropic.com/research/launc…
1
448
Jennifer Martinez retweeted
SITUATION CONFIRMED: every.to/vibe-check/fable-5-…
SITUATION DETECTED: Anthropic's Fable 5.1 is now live.
2
3
27
1,813
Jennifer Martinez retweeted
How might agentic coding tools, like Claude Code, alter the returns to expertise? In this report, we find evidence that domain expertise leads to better outcomes: more successful sessions and better recoveries after Claude struggles.
We studied seven months of Claude Code usage — for now, success on coding tasks depends little on whether you have a coding background … 1/
7
16
2
80
18,942
The vuln finding and fixing skills of Mythos get the most attention, but its autonomous exploitation skills are really where people should be focusing - this is why it marks such an immense turning point for cyber: axios.com/2026/06/08/exclusi…
205
Just the beginning
We’re expanding Project Glasswing. We’ve extended access to Claude Mythos Preview to approximately 150 additional organizations, based in more than fifteen countries. Read more about this expansion and our future plans for Project Glasswing: anthropic.com/news/expanding…
1
222
Jennifer Martinez retweeted
Claude Mythos absolutely destroys GPT-5.5 in ExploitBench and ExploitGym Mythos finds 18 arbitrary code execution exploits GPT-5.5 finds 0
I don't understand how people are still coping about Mythos. Here's a few benchmarks: SWE-bench Pro: Mythos -> 77.8%, GPT-5.5 -> 58.6% HLE: Mythos -> 56.8%, GPT-5.5 -> 41.4% UK AISI cyber ranges: - "The Last Ones": Mythos -> 6/10, GPT-5.5 3/10 - "Cooling Tower": Mythos -> 3/10, GPT-5.5 0/10 ExploitBench: - Mythos -> 18 Arbitrary Code Executions - GPT-5.5 -> 0 Arbitrary Code Executions ExploitGym: - Mythos -> 157 exploits (289.3 LLM calls) - GPT-5.5 -> 120 exploits (375.4 LLM calls) XBOW same story. Mythos has much higher odds of finding vulnerabilities within smaller token budgets.
73
70
9
1,131
135,913
This piece deserves attention - Mythos is the *first and only model**to solve this UK AISI cyber range (Cooling Tower). Also: When measuring time horizon tasks, Mythos Preview completes all six long tasks 100% of the time, **with a 2.5M token cap.** Other models do not.
Replying to @AISecurityInst
Mythos Preview also solved "Cooling Tower", our industrial control system range, in 3 of 10 attempts.
3
190
Jennifer Martinez retweeted
Two independent evals this week (XBOW and UK AISI) confirmed what my team has been seeing inside Project Glasswing: Claude Mythos Preview is a step change. The UK AISI analyzed the version of Mythos available at the launch of Project Glasswing and found it completed both of the AISI’s cyber ranges end-to-end, making it the first-ever model to do so! This is the start of an industry-wide response to address AI with powerful cyber capabilities. Planning to say more soon -- stay tuned!
A lot of people have been wondering about Mythos, Glasswing, and the vulns we / our partners are fixing. Today, I’m excited for us to start sharing more. (For context, I lead Glasswing @AnthropicAI.) Two independent evaluations this week—from XBOW and the UK AISI—confirm what we've been seeing internally: Claude Mythos Preview is a step change in autonomous cybersecurity capabilities. We need to start preparing fast for a world of models with this level of capabilities. The UK AI Security Institute tested the model we shipped at the launch of Project Glasswing and found Mythos Preview is the first model to solve both of their end-to-end cyber ranges, including one (Cooling Tower) which no model had ever cleared. But attackers (and defenders) have sophistication & cost constraints – Mythos is also the only model that clears every one of their tasks estimated over 8 hours under their deliberately low 2.5M-token cap. XBOW tested it on their offensive security benchmarks, finding "token-for-token, unprecedented precision." It's the only model to succeed at subtle V8 sandbox work. Other Glasswing partners shared similar stories. In a few weeks of testing, Mythos Preview has helped them find many thousands of (estimated) high + critical severity vulnerabilities, sometimes double what they'd normally find in a year. I don't share this to boost Mythos. In fact, this is not about Mythos. It’s about preparing for the coming world of models being better, faster, cheaper, and more creative than some of the best human experts at dual use capabilities. Clearly, we need them supporting defenders as widely as can be done safely – and especially the least resourced ones. Within a year, Mythos will probably look quite dumb (relative to other new models). And others may release openly available or unguardrailed models of Mythos-level capabilities. We started Project Glasswing because capabilities like Mythos Preview's won't stay rare, or stay in careful hands. We are bringing it to defenders as fast as we responsibly can, while working to figure out, for example, the right safeguards and patching & disclosure processes. Also, to be clear, compute has never been a limiter in our rollout. Expect a fuller update on our Glasswing work in the coming days. XBOW report: xbow.com/blog/mythos-offensi… UK AISI report: aisi.gov.uk/blog/how-fast-is…
4
2
27
2,460
Jennifer Martinez retweeted
The new version completely smashes GPT-5.5 and the previous Mythos version. Before Mythos Preview completed the cyber range 3 out of 10 times. The new version completed it 6 out of 10 times and is much more efficient!
Replying to @AISecurityInst
Our cyber range results illustrate this step-up. Since our first Mythos evaluation, we received access to a newer Mythos Preview checkpoint. On a 32-step corporate network attack we estimate takes a human expert ~20 hours, this checkpoint completes the full attack in 6 /10 attempts.
26
55
19
727
289,930
Jennifer Martinez retweeted
For the past 2 months, XBOW has been testing Mythos Preview under embargo as part of a select early-access group. Today, we can finally share what we found. The headline: Mythos Preview is a major advance. It is substantially better than prior models at finding vulnerability candidates, especially when source code is available. But it’s not perfect. We surfaced issues with exploit validation, judgment, and efficiency. Our full write-up covers where Mythos Preview shines, where it still needs support, and what we think this means for the future of offensive security: bit.ly/42zQl98
5
57
12
262
106,389
No one has quite dug into why Mythos is such a turning point like @nicoleperlroth did in this "Catch A Thief" episode with Anthropic security researcher Nicholas Carlini. If you think you've heard it all about Mythos and Project Glasswing, take a listen: youtube.com/watch?v=gKyV9gWZ…
2
165
Jennifer Martinez retweeted
Join me this Wednesday in SF for an event celebrating the new book from NPR's Planet Money team. We'll talk about the impact of AI on society, how we think about the future at Anthropic, and maybe read some of my Import AI writing. More info: cityarts.net/event/planet-mo…
6
7
1
60
11,877
Jennifer Martinez retweeted
Privileged to help lead this. Thankful to our partners. Mythos is an extraordinary model. But it is not about the model. It's about what the world needs to do to prepare for a future of models that are extremely good at cybersecurity. This is the start.
Introducing Project Glasswing: an urgent initiative to help secure the world’s most critical software. It’s powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans. anthropic.com/glasswing
51
49
7
1,065
130,521
Big. Congrats @logangraham and many others!
Introducing Project Glasswing: an urgent initiative to help secure the world’s most critical software. It’s powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans. anthropic.com/glasswing
5
168
Thanks for the discussion @ashleyrgold on our latest Economic Index report
Ashley Gold joined @AdrianaDiaz and @vladduthiersCBS to break down how AI fluency could shape America’s next class divide. 📺: @CBSNews | @ashleyrgold | @axios
6
171
Jennifer Martinez retweeted
New from the Anthropic Economic Index: how people’s use of Claude changes with experience. Longer-term users are more likely to iterate carefully with Claude, and less likely to hand it full autonomy. They attempt higher-value tasks, and receive more successful responses.
241
243
82
2,589
354,033
Jennifer Martinez retweeted
Jim VandeHei joined Morning Joe to discuss the latest on DHS and his and @mikeallen's column how AI fluency could shape America’s next class war. 📺: @Morning_Joe | @JimVandeHei | @axios
2
1
630