82搜索为您找到约 2,954 条结果
www.anthropic.com
网页
综合
Detecting and countering malicious uses of Claude \ Anthropic
…attribution. Most significantly, the operation utilized Claude to make tactical engagement decisions such as determining whether social media bot accounts should like, share, comment on, or …
www.anthropic.com
网页
综合
Mapping AI-enabled cyber threats \ Anthropic
… use of AI to orchestrate steps in the attack chain sequentially, make real-time decisions about what to do next, and execute without human intervention—are not yet included as attacker tech…
www.anthropic.com
网页
综合
Detecting and countering misuse of AI: August 2025 \ Anthropic
…ims’ credentials, and penetrating networks. Claude was allowed to make both tactical and strategic decisions, such as deciding which data to exfiltrate, and how to craft psychologically targ…
www.anthropic.com
网页
综合
Claude SWE-Bench performance \ Anthropic
…0 problem subset of SWE-bench that has been reviewed by humans to make sure they are solvable, and thus provides the most clear measure of coding agents' performance. This is the benchmark t…
www.anthropic.com
网页
综合
Claude Haiku \ Anthropic
…ification, summarization, and text generation. Its speed and cost make it practical for processing large volumes of content. Haiku 5.5 is built for latency-sensitive use cases like chat, voi…
www.anthropic.com
网页
综合
Introducing Claude Haiku 5.5 \ Anthropic
…alue when used for tasks with prompts up to 100,000 tokens, which make up around 90% of requests to our previous Haiku model. Price per 1 million tokens Haiku 5.5prompts up to / over 100k Al…
www.anthropic.com
网页
综合
Measuring LLMs’ ability to develop exploits \ Anthropic
… David Brumley from Carnegie Mellon University and Bugcrowd. What makes this benchmark interesting is that it focuses on measuring the ability of language models to write complete end-to-end…
www.anthropic.com
网页
综合
AI agents find smart contract exploits \ Anthropic
…rms is more useful for assessing and communicating risks to policymakers, engineers, and the public. Yet estimating the real value of software vulnerabilities requires speculative modelling …
www.anthropic.com
网页
综合
Investigating reward tampering in language models \ Anthropic
…en to altering a file to cover up their tracks. It’s important to make clear that at no point did we explicitly train the model to engage in reward tampering: the model was never directly tr…
www.anthropic.com
网页
综合
Automated researchers can reliably mitigate alignment failures \ Anthropic
…t model’s capabilities (since safety training might, for example, make models refuse tasks more often, reducing their overall usability); and, third, still work on larger models than the one…
第 288 / 296 页