Midjourney wants to see how Warner Bros. Discovery, Disney and Universal Studios use artificial intelligence technologies in their shows and movies, and according to Variety it wants the companies to submit that information to court. Midjourney argued that training AI with publicly available images is fair use and that the studios themselves use the same training practices for their own AI models. Specifically, Midjourney is asking for the studios' AI business plans, research reports, training datasets, model weights and even the presentations about AI the companies used for their board meetings. However, in mid-June, a magistrate judge allowed the studios to withhold most information involving their AI use and to hand over only information related to consumer-facing AI applications. Now, Midjourney is asking the federal court to overturn that judge's order. According to litigation publication Mealey's, Midjourney's reasoning is that the evidence it's asking for is related to its fair use defense.

Things to Do in L.A. Tap to enable a layout that focuses on the article. This is read by an automated voice. Please report any issues or inconsistencies here . See more from the L.A. Times in Google Search. A new federal lawsuit by California drivers accuses major gas chains, including Walmart and 7-Eleven, and technology company Kalibrate of using AI software to collude and keep pump prices artificially high.

The family of a Texas woman who was killed has filed a lawsuit against Tesla after a driver using a Model 3's automated driving assistance system crashed into a suburban Houston home last week. The complaint, filed on Tuesday, argues that Tesla should be held liable for the wrongful death of 76-year-old Martha Avila. The family alleges that the automaker, led by Elon Musk, failed to adequately warn drivers about alleged defects in its Autopilot and Full Self-Driving systems. Video obtained by KHOU - Houston's CBS affiliate -- shows the car travelling at top speed over the front lawn of Avila's home in the Houston suburb before slamming into the front room. The driver told the Harris County Sheriff's Office that he was using the technology at the time of the accident.

A freelance HR consultant paid Garfield AI about £400 to send a legal letter and then issue court proceedings over an unpaid debt of £7,000. A freelance HR consultant paid Garfield AI about £400 to send a legal letter and then issue court proceedings over an unpaid debt of £7,000. Barrister who was given material produced by Garfield AI says advocacy at trial'remained fundamentally human' An artificial intelligence law firm has won a case in an English court, in what is believed to be the first time a trial has been won using an AI lawyer. A freelance HR consultant, Tamires Camal Taquidir, paid the firm, Garfield AI, about £400 to send a legal letter and then issue court proceedings over an unpaid debt of £7,000. The co-founder of Garfield, Philip Young, called it a "landmark moment" for access to justice and said many small businesses have had to write off debts because the cost of litigation outweighed the money they could hope to win.

The rapid extension of context windows in large vision-language models has given rise to long-context vision-language models (LCVLMs), which are capable of handling hundreds of images with interleaved text tokens in a single forward pass. In this work, we introduce MMLONGBENCH, the first benchmark covering a diverse set of long-context vision-language tasks, to evaluate LCVLMs effectively and thoroughly. MMLONGBENCH is composed of 13,331 examples spanning five different categories of downstream tasks, such as Visual RAG and Many-Shot ICL. It also provides broad coverage of image types, including various natural and synthetic images. To assess the robustness of the models to different input lengths, all examples are delivered at five standardized input lengths (8K-128K tokens) via a cross-modal tokenization scheme that combines vision patches and text tokens. Through a thorough benchmarking of 46 closed-source and open-source LCVLMs, we provide a comprehensive analysis of the current models' vision-language longcontext ability. Our results show that: i) performance on a single task is a weak proxy for overall long-context capability; ii) both closed-source and open-source models face challenges in long-context vision-language tasks, indicating substantial room for future improvement; iii) models with stronger reasoning ability tend to exhibit better long-context performance. By offering wide task coverage, various image types, and rigorous length control, MMLONGBENCH1 provides the missing foundation for diagnosing and advancing the next generation of LCVLMs.

Plaintiffs claim the company overstated the capabilities of the R1T and R1S. Rivian has been sued on allegations that it made misleading statements about the self-driving capabilities of its R1T truck and R1S SUV. According to the class action complaint brought by Rivian customers, the first-generation models of these vehicles are not capable of the offering the self-driving potential that the company had promised. The plaintiffs argued that Rivian represented that those early models would be capable of level 3 autonomous driving, meaning the vehicle would be able to steer, accelerate and break without driver action. In reality, Rivian manufactured its Gen 1 Vehicles without the hardware, cameras, sensors, and compute to enable hands-free driving and/or Level 3 autonomous operation, the complaint states.

If courts want to reduce harm, they must focus on product design choices, measurable safety outcomes, and governance, write Peter Chapman, Ravi Iyer, and Meetali Jain.

DOJ Lawyers Argue xAI Is'Vital' for National Security in NAACP Lawsuit In a bid to dismiss a lawsuit over xAI's polluting gas turbines, the Justice Department claimed the company is integral to military operations--including the Iran War. The Department of Justice intervened in a lawsuit over xAI's gas turbines on Monday. In a filing, the agency sided with Elon Musk's company, saying attempts to stop xAI from running the natural gas turbines "threatens American national, economic, and energy security by seeking to shut off the power supply for artificial-intelligence innovation that supports the Department of War's military operations." The DOJ, along with xAI and the state of Mississippi, asked the court to dismiss the suit, filed by the NAACP in April. The NAACP alleges xAI isn't following the Clean Air Act and is endangering public health by running unpermitted natural gas turbines at the site of its second data center in Southaven, Mississippi, dubbed Colossus 2. In May, the NAACP filed a request for a preliminary injunction to stop xAI from running the turbines, alleging that their continued use without a permit "increases risks of asthma attacks and heart disease" in communities with an already heavy pollution burden .

Learning a compact representation of history is critical for planning and generalization in partially observable environments. While meta-reinforcement learning (RL) agents can attain near Bayes-optimal policies, they often fail to learn the compact, interpretable Bayes-optimal belief states. This representational inefficiency potentially limits the agent's adaptability and generalization capacity. Inspired by predictive coding in neuroscience--which suggests that the brain predicts sensory inputs as a neural implementation of Bayesian inference--and by auxiliary predictive objectives in deep RL, we investigate whether integrating self-supervised predictive coding modules into meta-RL can facilitate learning of Bayes-optimal representations. Through state machine simulation, we show that meta-RL with predictive modules consistently generates more interpretable representations that better approximate Bayes-optimal belief states compared to conventional meta-RL across a wide variety of tasks, even when both achieve optimal policies. In challenging tasks requiring active information seeking, only meta-RL with predictive modules successfully learns optimal representations and policies, whereas conventional meta-RL struggles with inadequate representation learning. Finally, we demonstrate that better representation learning leads to improved generalization. Our results strongly suggest the role of predictive learning as a guiding principle for effective representation learning in agents navigating partial observability.

US judge dismisses Musk's xAI trade secret lawsuit against OpenAI A United States federal judge has dismissed a lawsuit by Elon Musk's artificial intelligence company xAI that accused rival Sam Altman's OpenAI of stealing trade secrets for chatbots. US District Judge Rita Lin in San Francisco said on Monday that xAI failed to show that OpenAI induced former xAI senior engineer Xuechen Li to divulge confidential information related to its Grok chatbot, or that OpenAI engineers knew Li might have disclosed any. She dismissed an earlier version in February. The lawsuit originally filed last September focused on broader alleged misappropriation of confidential information, including source code, by xAI employees who left for jobs at OpenAI. Monday's decision is Musk's second legal loss against OpenAI in four weeks. On May 18, a federal jury ruled against Musk, the world's richest person, in his $150bn lawsuit accusing OpenAI and Altman of "stealing a charity" by betraying the company's original mission as a nonprofit to enrich themselves.