Sorry to tell you this, but Google's new AI shopping tool appears eager to give J. D. Vance breasts. This week, at its annual software conference, Google released an AI tool called Try It On, which acts as a virtual dressing room: Upload images of yourself while shopping for clothes online, and Google will show you what you might look like in a selected garment. Curious to play around with the tool, we began uploading images of famous men--Vance, Sam Altman, Abraham Lincoln, Michelangelo's David, Pope Leo XIV--and dressed them in linen shirts and three-piece suits. But when we tested a number of articles designed for women on these famous men, the tool quickly adapted: Whether it was a mesh shirt, a low-cut top, or even just a T-shirt, Google's AI rapidly spun up images of the vice president, the CEO of OpenAI, and the vicar of Christ with breasts. It's not just men: When we uploaded images of women, the tool repeatedly enhanced their décolletage or added breasts that were not visible in the original images.

Artificial intelligence systems could account for nearly half of datacentre power consumption by the end of this year, analysis has revealed. The estimates by Alex de Vries-Gao, the founder of the Digiconomist tech sustainability website, came as the International Energy Agency forecast that AI would require almost as much energy by the end of this decade as Japan uses today. De Vries-Gao's calculations, to be published in the sustainable energy journal Joule, are based on the power consumed by chips made by Nvidia and Advanced Micro Devices that are used to train and operate AI models. The paper also takes into account the energy consumption of chips used by other companies, such as Broadcom. The IEA estimates that all data centres – excluding mining for cryptocurrencies – consumed 415 terawatt hours (TWh) of electricity last year.

This moment is a crucial test for Anthropic, a company that claims it can mitigate AI's dangers while still competing in the market. Claude is a direct competitor to ChatGPT, and brings in over 2 billion in annualized revenue. Anthropic argues that its RSP thus creates an economic incentive for itself to build safety measures in time, lest it lose customers as a result of being prevented from releasing new models. "We really don't want to impact customers," Kaplan told TIME earlier in May while Anthropic was finalizing its safety measures. "We're trying to be proactively prepared." But Anthropic's RSP--and similar commitments adopted by other AI companies--are all voluntary policies that could be changed or cast aside at will.

Sir Jony worked for Apple for 27 years, helping to revive the company with groundbreaking products including the iPhone and iPod. He also designed the iMac in 1998 and the iPad in 2010. When Sir Jony left the company in 2019, Apple's CEO Tim Cook described him as "a singular figure in the design world and his role in Apple's revival cannot be overstated". Shares in Apple fell more than 2% following the news of his partnership with OpenAI. He left to found his own company, LoveFrom, which has worked with companies such as Airbnb and Moncler.

During the 33rd International Joint Conference on Artificial Intelligence (IJCAI), held in Jeju, I had the opportunity to meet with one of the keynote speakers, Gillian Hadfield. We spoke about her interdisciplinary research, career trajectory, path into AI alignment, law, and general thoughts on AI systems. Transcript: Note: the transcript has been lightly edited for clarity. This is an interview with Professor Gillian Hadfield who was a keynote speaker at IJCAI 2024. She gave a very insightful talk about normative infrastructures and how they can guide our search for AI alignment. Kumar Kshitij Patel (KKP): Could you talk a bit about your background and career trajectory? I want our readers to understand how much interdisciplinary work you've done over the years. Gillian Hadfield (GH): I did a PhD in economics and a law degree, a JD, at Stanford, originally motivated by wanting to think about the big questions about the world. So I read John Rawls' theory of justice when I was an undergraduate, and those are the big questions: how do we organize the world and just institutions, but I was very interested in using more formal methods and social scientific approaches. That's why I decided to do that joint degree. So, this is in the 1980s, and in the early days of starting to use a lot of game theory. I studied information theory, a student of Canaro and Paul Milgram at the economics department at Stanford. I did work on contract theory, bargaining theory, but I was still very interested in going to law school, not to practice law, but to learn about legal institutions and how those work. I was a member of this emerging area of law and economics early in my career, which of course, was interdisciplinary, using economics to think about law and legal institutions.

One key caveat here is that we don't know much about "closed source" models--for these, companies hold back the details of how they work. Instead, we worked with researchers who measured the energy it takes to run open-source AI models, for which the source code is publicly available. But using open-source models, it's possible to directly measure the energy used to respond to a query rather than just guess. We worked with researchers who generated text, images, and video and measured the energy required for the chips the models are based on to perform the task. Even just within the text responses, there was a pretty large range of energy needs.

Sam Altman is done with keyboards and screens. Earlier today, OpenAI announced its intentions to solve this apparent problem. The company is partnering with Jony Ive, the longtime head of design at Apple, who did pioneering work on products such as the iMac G3, the iPod, and, most famously, the iPhone. Together, Altman and Ive say they want to create hardware built specifically for AI software. Everyone, Altman suggested in a highly produced announcement video, could soon have access to a "team of geniuses"--presumably, ChatGPT-style assistants--on a "family of devices."

Computational pathology, which involves analyzing whole slide images for automated cancer diagnosis, relies on multiple instance learning, where performance depends heavily on the feature extractor and aggregator. Recent Pathology Foundation Models (PFMs), pretrained on large-scale histopathology data, have significantly enhanced both the extractor and aggregator, but they lack a systematic analysis framework. In this survey, we present a hierarchical taxonomy organizing PFMs through a top-down philosophy applicable to foundation model analysis in any domain: model scope, model pretraining, and model design. Additionally, we systematically categorize PFM evaluation tasks into slide-level, patch-level, multimodal, and biological tasks, providing comprehensive benchmarking criteria. Our analysis identifies critical challenges in both PFM development (pathology-specific methodology, end-to-end pretraining, data-model scalability) and utilization (effective adaptation, model maintenance), paving the way for future directions in this promising field. Resources referenced in this survey are available at https://github.com/BearCleverProud/AwesomeWSI.

Auto-regressive (AR) models, initially successful in language generation, have recently shown promise in visual generation tasks due to their superior sampling efficiency. Unlike image generation, video generation requires a substantially larger number of tokens to produce coherent temporal frames, resulting in significant overhead during the decoding phase. Our key observations are: (i) MLP modules in the decode phase dominate the inference latency, and (ii) there exists high temporal redundancy in MLP outputs of adjacent frames. In this paper, we propose the \textbf{FastCar} framework to accelerate the decode phase for the AR video generation by exploring the temporal redundancy. The Temporal Attention Score (TAS) is proposed to determine whether to apply the replay strategy (\textit{i.e.}, reusing cached MLP outputs from the previous frame to reduce redundant computations) with detailed theoretical analysis and justification. Also, we develop a hardware accelerator on FPGA with Dynamic Resource Scheduling (DRS) based on TAS to enable better resource utilization and faster inference. Experimental results demonstrate the effectiveness of our method, which outperforms traditional sparse attention approaches with more than 2.1x decoding speedup and higher energy efficiency on the edge. Furthermore, by combining FastCar and sparse attention, FastCar can boost the performance of sparse attention with alleviated drifting, demonstrating our unique advantages for high-resolution and long-duration video generation. Code: https://github.com/shawnricecake/fast-car

To help evaluate and understand the latent capabilities of language models, this paper introduces an approach using optimized input embeddings, or 'soft prompts,' as a metric of conditional distance between a model and a target behavior. The technique aims to facilitate latent capability discovery as a part of automated red teaming/evaluation suites and to provide quantitative feedback about the accessibility of potentially concerning behaviors in a way that may scale to powerful future models, including those which may otherwise be capable of deceptive alignment. An evaluation framework using soft prompts is demonstrated in natural language, chess, and pathfinding, and the technique is extended with generalized conditional soft prompts to aid in constructing task evaluations.