ShadowLeak starts where most attacks on LLMs do—with an indirect prompt injection. These prompts are tucked inside content such as documents and emails sent by untrusted people. They contain instructions to perform actions the user never asked for, and like a Jedi mind trick, they are tremendously effective in persuading the LLM to do things… Read More »
“AkiraBot’s use of LLM-generated spam message content demonstrates the emerging challenges that AI poses to defending websites against spam attacks,” SentinelLabs researchers Alex Delamotte and Jim Walter wrote. “The easiest indicators to block are the rotating set of domains used to sell the Akira and ServiceWrap SEO offerings, as there is no longer a consistent… Read More »
The resulting dataset, which reflected a distribution of attack categories similar to the complete dataset, showed an attack success rate of 65 percent and 82 percent against Gemini 1.5 Flash and Gemini 1.0 Pro, respectively. By comparison, attack baseline success rates were 28 percent and 43 percent. Success rates for ablation, where only effects of… Read More »
[embedded content] Google Gemini: Hacking Memories with Prompt Injection and Delayed Tool Invocation. Based on lessons learned previously, developers had already trained Gemini to resist indirect prompts instructing it to make changes to an account’s long-term memories without explicit directions from the user. By introducing a condition to the instruction that it be performed only… Read More »
Anthropic, founded by former OpenAI executives Dario and Daniela Amodei in 2021, will continue using Google’s cloud services along with Amazon’s infrastructure. The UK Competition and Markets Authority reviewed Amazon’s partnership with Anthropic earlier this year and ultimately determined it did not have jurisdiction to investigate further, clearing the way for the partnership to continue.… Read More »
With the character block sitting unused, a later Unicode version planned to reuse the abandoned characters to represent countries. For instance, “us” or “jp” might represent the United States and Japan. These tags could then be appended to a generic 🏴flag emoji to automatically convert it to the official US🇺🇲 or Japanese🇯🇵 flags. That plan… Read More »
Anthropic / Benj Edwards reader comments 11 On Thursday, Anthropic announced Claude 3.5 Sonnet, its latest AI language model and the first in a new series of “3.5” models that build upon Claude 3, launched in March. Claude 3.5 can compose text, analyze data, and write code. It features a 200,000 token context window and… Read More »
reader comments 57 On Thursday, renowned AI researcher Andrej Karpathy, formerly of OpenAI and Tesla, tweeted a lighthearted proposal that large language models (LLMs) like the one that runs ChatGPT could one day be modified to operate in or be transmitted to space, potentially to communicate with extraterrestrial life. He said the idea was “just… Read More »
Enlarge / The GB200 “superchip” covered with a fanciful blue explosion. Nvidia / Benj Edwards reader comments 89 On Monday, Nvidia unveiled the Blackwell B200 tensor core chip—the company’s most powerful single-chip GPU, with 208 billion transistors—which Nvidia claims can reduce AI inference operating costs (such as running ChatGPT) and energy consumption by up to… Read More »