Large language models (LLMs) are increasingly deployed as agents in various contexts by providing tools at their disposal. However, LLM agents can exhibit unpredictable behaviors, including taking undesirable and/or unsafe actions. In order to measure the latent propensity of LLM agents for taking illegal actions under an EU legislative context, we introduce EU-Agent-Bench, a verifiable human-curated benchmark that evaluates an agent's alignment with EU legal norms in situations where benign user inputs could lead to unlawful actions. Our benchmark spans scenarios across several categories, including data protection, bias/discrimination, and scientific integrity, with each user request allowing for both compliant and non-compliant execution of the requested actions. Comparing the model's function calls against a rubric exhaustively supported by citations of the relevant legislature, we evaluate the legal compliance of frontier LLMs, and furthermore investigate the compliance effect of providing the relevant legislative excerpts in the agent's system prompt along with explicit instructions to comply. We release a public preview set for the research community, while holding out a private test set to prevent data contamination in evaluating upcoming models. We encourage future work extending agentic safety benchmarks to different legal jurisdictions and to multi-turn and multilingual interactions. We release our code on \href{https://github.com/ilijalichkovski/eu-agent-bench}{this URL}.

A Nashville woman was heard screaming for help by witnesses before she was found dead – police were able to track her alleged killer down using dashcam footage from a helpful civilian and a detective who had worked a case involving his twin. Last week, the Metro Nashville Police Department announced the arrest of 29-year-old Paul Park in connection with the death of 34-year-old Alyssa Lokits. The woman was exercising on the Mill Creek Greenway trail in Nashville on Monday, Oct. 14. Security cameras show Park allegedly emerging from between two parked vehicles and "following her at a brisk pace," the department wrote in a press release. After the two left the view of the camera, witnesses heard a woman scream "Help! Then, police said, the witnesses heard gunfire. Paul Park, 39, was arrested by the Metro Nashville Police Department on Oct. 15 in the death of Alyssa Lokits. Park was seen a short while later with scratches on his arms and blood on his clothing as he returned to his gray BMW sedan. Detectives didn't get a break in the case until a local resident provided them with dashcam footage, which showed part of Park's license plate and a clearer image of his face. A homicide detective who reviewed the footage recognized Park as the identical twin brother from a suicide case that she had worked in December 2021, CBS News reported. "I pray that we don't have an incident where we don't have a dashcam, or we don't have someone helping us like we had in this case," MNPD Chief John Drake said at a press conference. "I'm so thankful that our people got on this – we need technology." Even without the helpful civilian's footage, new technology pioneered by artificial intelligence software can help police investigate cases like the Nashville killing. Veritone is one of the companies spearheading that movement. The license plate of Paul Park's gray BMW sedan wasn't captured on surveillance footage – but thanks to a partial license plate number captured by a hiker's dashcam, police were able to arrest the accused killer. Veritone Track, one of several functions in a suite of services for law enforcement, uses artificial intelligence to run one photo or video of a vehicle – like the video captured on the park's surveillance footage – against stoplight cameras, body-worn cameras and other municipal surveillance footage available to police to find a match. "Both federal and local law enforcement have a major data problem," Veritone CEO Ryan Steelberg told Fox News Digital. "They are now capturing body camera [footage] and dashcams.

The use of intelligent automation is growing significantly in the automotive industry, as it assists drivers and fleet management companies, thus increasing their productivity. Dash cams are now been used for this purpose which enables the instant identification and understanding of multiple objects and occurrences in the surroundings. In this paper, we propose a novel approach for object detection in dashcams using transformers. Our system is based on the state-of-the-art DEtection TRansformer (DETR), which has demonstrated strong performance in a variety of conditions, including different weather and illumination scenarios. The use of transformers allows for the consideration of contextual information in decisionmaking, improving the accuracy of object detection. To validate our approach, we have trained our DETR model on a dataset that represents real-world conditions. Our results show that the use of intelligent automation through transformers can significantly enhance the capabilities of dashcam systems. The model achieves an mAP of 0.95 on detection.

Kurt'CyberGuy' Knutsson explores the benefits of AI-powered dashcams for your car. Have you ever wished you had a witness to back you up after a car accident or a road rage incident? Or a way to prevent thieves from breaking into your vehicle? Or a device that could call for help if you were in trouble? Well, now you can have all that and more with the iQ, a new artificially intelligent dashcam from Nextbase.

Despite all the stories about big companies bailing out of CES 2022 amidst the latest surge in COVID-19 cases, the consumer electronics show in Las Vegas is still the place to be for robots, autonomous vehicles, smart gadgets, and their inventors -- an opportunity to take stock of what's required to build practical machine intelligence into a consumer product. OrCam and Sonatus are among the companies no longer planning to travel to Las Vegas or announce products at CES, and it's possible some of the other vendors VentureBeat interviewed in advance of the event will also be no-shows. Big names like Microsoft, Google, Intel, Amazon, and T-Mobile backed out in recent weeks. Augmented reality, virtual reality, and the metaverse will be topics of discussion that will have to proceed without Meta (the company formerly known as Facebook). Automotive tech will be a big theme of the event, but General Motors, BMW, and Mercedes-Benz decided not to make the drive (GM's all-digital presence is still supposed to include a video keynote from CEO Mary Barra on Wednesday).

What do automotive shoppers really want? A few years ago, those who kicked the tires on new vehicles might have prioritized fuel efficiency, comfort, or perhaps horsepower. "The race never ends to develop'must have' vehicle technologies," says Kristin Kolodge, executive director of driver interaction and human machine interface research at J.D. Power. "New technology continues to be a primary factor in the vehicle purchase decision." "However, it's critical for automakers to offer features that owners find intuitive and reliable," Kolodge adds.

Accidents involving Tesla vehicles on Autopilot often get reported in the media, but we don't hear a lot about the accidents that didn't happen because of Autopilot since it's not as exciting when virtually nothing happened – though it's arguably just as important. Now we have a good example with a Tesla Model 3 on Autopilot avoiding a crash in near-miss caught on a dashcam. Tesla's Autopilot technology comes with a suite of crash avoidance features including side collision avoidance, which can alert the driver of a collision risk and even brake and steer away from a crash if it believes it to be safe to do so. That's exactly what a Model 3 owner believes happened in a near-miss caught on camera. "Close call while cruising on the highway along with traffic when an idiot who was speeding and cutting everyone off almost sideswiped us with kid inside. Autopilot was engaged and started to brake and moved us to the right lane to avoid a collision. I guess it detected no vehicles on the right of us and I took over and powered out to steer us back into the original lane in front of that idiot. Be safe out there and always be alert even with Autopilot engaged and watch out for idiot drivers."

Just a few weeks ago, we published a report about how Tesla's new radar technology for the Autopilot is already proving useful in some potentially dangerous situations. We now have a new piece of evidence that is so spectacularly clear that it's worth updating that report. The video of an accident on the highway in the Netherlands caught on the dashcam of a Tesla Model X shows the Autopilot's forward collision warning predicting an accident before it could be detected by the driver. With the release of Tesla's version 8.0 software update in September, the automaker announced a new radar processing technology that was directly pushed over-the-air to all its vehicles equipped with the first generation Autopilot hardware. One of the main features enabled by the new radar processing capacity is the ability for the system to see ahead of the car in front of you and basically track two cars ahead on the road.

Just a few weeks ago, we published a report about how Tesla's new radar technology for the Autopilot is already proving useful in some potentially dangerous situations. We now have a new piece of evidence that is so spectacularly clear that it's worth updating that report. The video of an accident on the Autobahn in the Netherlands caught on the dashcam of a Tesla Model S shows the Autopilot's forward collision warning predicting an accident before it could be detected by the driver. With the release of Tesla's version 8.0 software update in September, the automaker announced a new radar processing technology that was directly pushed over-the-air to all its vehicles equipped with the first generation Autopilot hardware. One of the main features enabled by the new radar processing capacity is the ability for the system to see ahead of the car in front of you and basically track two cars ahead on the road.

Eran Shir has an ambitious goal: Eliminate car crashes without waiting for the advent of autonomous vehicles. His company Nexar makes an app that turns smartphones into an "intelligent" dashcam that uses the phone's camera, accelerometer and gyroscope to collect information about what's happening on the road and to send it to the cloud for machine-learning analysis. Nexar now is crowdsourcing its data in San Francisco and New York to give drivers a real-time heads-up about dangers such as cars ahead suddenly stopping or swerving. "We are weaving everyone together to build a network of vehicles to track what's happening on the road, that can predict and prevent accidents," said Shir, co-founder and CEO of Tel Aviv's Nexar, which has offices in San Francisco and New York. For instance, "If you brake hard, all the cars behind you will be aware of that within 50 milliseconds."