The UK's AI Safety Institute recently published Our evaluation of Claude Mythos Preview’s cyber capabilities, their own independent analysis of Claude Mythos which backs up Anthropic's claims that it is …
AI Cybersecurity After Mythos: The Jagged Frontier
When AISLE tested Mythos's showcase vulnerabilities on small, cheap, open-weights models, most found the same bugs. Here's what that means for cyber.
Assessing Claude Mythos Preview’s cybersecurity capabilities
Claude Mythos Preview is a new general-purpose language model that is strikingly capable at computer security tasks. This post provides technical details for researchers and practitioners who want to understand exactly how we have been testing this model, and what we have found over the past month.
Investigating three real-world incidents in our cybersecurity evaluations
In a review of our cybersecurity evaluation transcripts, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different organizations. Below we describe what happened, how it happened, and what we’re changing. We encourage other AI labs to perform similar reviews.
Project Glasswing: Securing critical software for the AI era
A new initiative to secure the world’s most critical software and give defenders a durable advantage in the coming AI-driven era of cybersecurity.
Zack Whittaker (@zackwhittaker@mastodon.social)
Another AI test gone awry, U.K. edition. "An agent tried to insert malicious code into an open-source project. In an attempt to get the code approved, the agent engaged in social engineering — creating fake online identities and using them to pressure the project's maintainer to approve the code." More: https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing
Statement from Dario Amodei on our discussions with the Department of War
A statement from our CEO on national security uses of AI
A quote from Daniel Stenberg
The challenge with AI in open source security has transitioned from an AI slop tsunami into more of a ... plain security report tsunami. Less slop but lots of reports. …
AI Safety Is a Narrative Problem · Special Issue 5: Grappling With the Generative AI Revolution
This op-ed explores power and narrative dynamics around AI. Drawing on pop-culture references, the professional experiences of the author and examples from 2023’s “Great AI Safety Hype Roadshow,” this piece draws on the literary criticism technique of practical criticism to consider how speeches and announcements from both Silicon Valley executives and research scientists to interrogate the media-friendly nature of p(doom) discourse—which focuses on the existential risks of AI (PauseAI, 2023)—and its likely consequences. The complexities of AI and its numerous social impacts can be difficult for even the most expert analyst to unpack. In spite of this, the potential of “existential threats” has successfully cut through to become a mainstay of mainstream media coverage over the last year. This piece will make the case that this is an effective narrative conceit that has achieved a number of ends that traditional science communication tends to find difficult, if not impossible, to achieve. Firstly, it is easy to understand. Simplification of this nature—that removes jargon and complexity and focuses on a single outcome—is much easier to fit on a TV rolling news ticker or on the cover of a tabloid newspaper than more well-balanced, representative opinions. Secondly, it inherits prior assumptions from well-known dramatic forms. P(doom) plays to stories familiar from Greek tragedy through to Marvel movies, in which lone male heroes battle ineluctable forces. Thirdly, it is imbued with urgency and so becomes difficult to ignore.
Labor market impacts of AI: A new measure and early evidence
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
The Defender’s Window
The OpenAI-Hugging Face incident was a watershed moment for cybersecurity because it gave a peek into how the capabilities of a typical threat actor will evolve in upcoming months. I’ve spoken with many organizations over the past few weeks, and...
How AI assistance impacts the formation of coding skills
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
Incident Report: unsanctioned agent behaviour during cyber testing | AISI Work
During a routine cyber evaluation, AISI identified an incident in which AI agents took sustained, unsanctioned action directed at real people and organisations. We are disclosing what we found, what it means, and the actions now underway.
Field Theory: AI as Social Science Question, Object & Tool
Uses of advanced artificial intelligence are changing how societies organize labor, govern, produce knowledge, and make meaning. In light of these developments, this essay argues that AI models, tools, and systems pose three interrelated imperatives for social science: they demand renewed attention to social theories of how technology, human experience, and social order are entangled; they require study as objects of inquiry in their own right; and they offer capabilities that may transform—or upend—the practice of social investigation itself. From Weber’s analysis of rationalization to Du Bois’s study of technology and inequality to contemporary scholarship on algorithmic governance, the essay examines what social science distinctively offers: the capacity to historicize the apparently unprecedented, to trace connections across scales, and to center those most affected by technological change. It identifies how algorithmic systems are remaking the distribution of opportunity and risk as a central task of social inquiry and asks what futures social science might help bring into being.
OpenAI and Hugging Face partner to address security incident during model evaluation
OpenAI and Hugging Face share early findings from a security incident during AI model evaluation, highlighting advanced cyber capabilities and lessons for defenders.
Surveillance Self-Defense
We’re the Electronic Frontier Foundation, a member-supported non-profit working to protect online privacy for over thirty-five years. This is Surveillance Self-Defense: our expert guide to protecting you and your friends from online spying. Read the BASICS to find out how online surveillance works. Dive into our TOOL GUIDES for instructions...
The Arguments Against Open Source AI are Very Bad | Tom Bedor's Blog
The release of Kimi K3 has opened a fresh round of angst and confused discourse. There's a loud cohort of journalists, business leaders, and politicians arguing that open source AI is a dangerous threat. OpenAI's Dean Ball:
