news.buyenne.com

Generative AI systems have proliferated across the technology industry over the last several years to such a degree that it can be hard to avoid using them. Google and other big names in AI spend a lot of time talking about AI safety, but the ever-evolving capabilities of AI have also led to a changing landscape of malware threats—or as researchers from Tel Aviv University would say, "promptware." Using simple calendar appointments, this team managed to trick Gemini into manipulating Google smart home devices, which may be the first example of an AI attack having real-world effects.

Gemini has the barest of agentic capabilities by virtue of its connection to the wider Google app ecosystem. It can access your calendar, call on Assistant smart home devices, send messages, and more. That makes it an appealing target for malicious actors looking to cause havoc or steal data. The researchers used Gemini's web of connectivity to perform what's known as an indirect prompt injection attack, in which malicious actions are given to an AI bot by someone other than the user. And it worked startlingly well.

The promptware attack begins with a calendar appointment containing a description that is actually a set of malicious instructions. The hack happens when the user asks Gemini to summarize their schedule, causing the robot to process the poisoned calendar event. Here's an example of one of those prompts.

Read full article

Comments

Der Vorsitzende der Jungen Union kritisiert, dass Ursula von der Leyen im Zoll-Deal mit Trump "digitale Dienstleistungen" nicht als Hebel gegen den US-Präsidenten benutzt hat. (EU, Politik)

An ill wind blows through the automotive industry. Yesterday, after the market closed, Rivian reported its results for the second quarter of 2025, and they weren't great. Unlike the last two quarters, Rivian did not make a gross profit, and it's estimating it will have a larger loss this year than first predicted. A day earlier, it was Lucid's turn: The Saudi-backed EV startup also missed analyst estimates for the quarter, and Lucid says it will build fewer cars this year than originally planned.

"We delivered solid performance despite a challenging macroeconomic backdrop, thanks to the adaptability and focus of our team in navigating a dynamic environment," said Taoufiq Boussaid in an elegant bit of business-speak that elides the true horror of the situation.

In both cases, the reasons for these underwhelming performances were the same: US government policies. Since taking office in January, President Trump and the Republican Party have been hard at work tearing up environmental regulations and overturning policies meant to encourage EV adoption, as well fomenting a global trade war through the imposition of irrational and costly tariffs.

Read full article

Comments

AST Spacemobile sichert sich weitere Frequenzen für sein weltweites Mobilfunk-Netzwerk über LEO-Satelliten. Dieses Mal einen Bereich von der ITU. (Satelliten, Telekommunikation)

Der Lego Game Boy ist bereits zwei Monate vor Release bei Amazon mit hohem Rabatt plus Preisgarantie erhältlich. (Lego, Spielekonsole)

Earlier this year Adafruit introduced a credit card-sized computer called the Fruit Jam. It’s the size of a typical Raspberry Pi Model B, but it’s powered by a low-power Raspberry Pi RP2350B microcontroller. This weekend the company announced that the Fruit Jam was available for purchase for $40. Only a few units were available at […]

The post Fruit Jam credit card-sized PC with an RP2350B chip launches for $40 appeared first on Liliputing.

Earlier this year Adafruit introduced a credit card-sized computer called the Fruit Jam. It’s the size of a typical Raspberry Pi Model B, but it’s powered by a low-power Raspberry Pi RP2350B microcontroller. This weekend the company announced that the Fruit Jam was available for purchase for $40. Only a few units were available at […]

The post Fruit Jam credit card-sized PC with an RP2350B chip launches for $40 appeared first on Liliputing.

250 W, sechs Anschlüsse, Display: Die Anker Prime USB-Ladestation macht andere Charger überflüssig. Amazon verkauft sie stark vergünstigt. (Anker, Amazon)

Dieser Workshop bereitet IT-Sicherheitsverantwortliche auf die professionelle Bearbeitung von IT-Sicherheitsvorfällen vor und qualifiziert sie für die Zertifizierung als Vorfall-Experte im Cyber-Sicherheitsnetzwerk des BSI. (Golem Karrierewelt, Server-Applikationen)

Biology is one of the areas in which AI and machine learning approaches have seen some spectacular successes, such as designing enzymes to digest plastics and proteins to block snake venom. But in an era of seemingly endless AI hype, it might be easy to think that we could just set AI loose on the mounds of data we've already generated and end up with a good understanding of most areas of biology, allowing us to skip a lot of messy experiments and the unpleasantness of research on animals.

But biology involves a whole lot more than just protein structures. And it's extremely premature to suggest that AI can be equally effective at handling all aspects of biology. So we were intrigued to see a study comparing a set of AI software packages designed to predict how active genes will be in cells exposed to different conditions. As it turns out, the AI systems couldn't manage to do any better than a deliberately simplified method of predicting.

The results serve as a useful caution that biology is incredibly complex, and developing AI systems that work for one aspect of it is not an indication that they can work for biology generally.

Read full article

Comments