AI chatbots are giving out people's real phone numbers People report that their personal contact info was surfaced by Google AI--and there's apparently no easy way to prevent it. A Redditor recently wrote that he was "desperate for help": for about a month, he said, his phone had been inundated by calls from "strangers" who were "looking for a lawyer, a product designer, a locksmith." Callers were apparently misdirected by Google's generative AI. In March, a software developer in Israel was contacted on WhatsApp after Google's chatbot Gemini provided incorrect customer service instructions that included his number. And in April, a PhD candidate at the University of Washington was messing around on Gemini and got it to cough up her colleague's personal cell phone number. AI researchers and online privacy experts have long warned of the myriad dangers generative AI poses for personal privacy.

People who are screened for breast cancer by AI-supported radiologists are less likely to develop aggressive cancers before their next screening round than those who are screened by radiologists alone, raising hopes that AI-assisted screening could save lives. "This is the first randomised controlled trial on the use of AI in mammography screening," says Kristina Lång at Lund University in Sweden. The AI-supported approach involves using the software - which has been trained on more than 200,000 mammography scans from 10 countries - to rank the likelihood of cancer being present in mammograms on a scale of 1 to 10, based on visual patterns in the scans. The scans receiving a score of 1 to 9 are then assessed by one experienced radiologist, while scans receiving a score of 10 - indicating cancer is most likely to be present - are assessed by two experienced radiologists. An earlier study found that this approach could detect 29 per cent more cancers than standard screening, where each mammogram is assessed by two radiologists, without increasing the rate of false detections - where a growth is flagged but follow-up tests reveal it isn't actually there or wouldn't go on to cause problems.

In August, X, the social media company once known as Twitter, publicly released Grok 2, the latest iteration of its AI chatbot. With limited guardrails, Grok has been responsible for pushing misinformation about elections and allowing users to make life-like artificial intelligence-generated images – otherwise known as deepfakes – of elected officials in ethically questionable positions. The social media giant has started to rectify some of its problems. After election officials in Michigan, Minnesota, New Mexico, Pennsylvania and Washington wrote to X head Elon Musk alleging that the chatbot produced false information about state ballot deadlines, X now points users to Vote.gov for election-related questions. But when it comes to deepfakes, that's a different story.

The AI election is here. Already this year, a robocall generated using artificial intelligence targeted New Hampshire voters in the January primary, purporting to be President Joe Biden and telling them to stay home in what officials said could be the first attempt at using AI to interfere with a US election. The "deepfake" calls were linked to two Texas companies, Life Corporation and Lingo Telecom. It's not clear if the deepfake calls actually prevented voters from turning out, but that doesn't really matter, said Lisa Gilbert, executive vice-president of Public Citizen, a group that's been pushing for federal and state regulation of AI's use in politics. "I don't think we need to wait to see how many people got deceived to understand that that was the point," Gilbert said.

The British government should offer tax breaks for businesses developing AI-powered products and services, or applying AI to their existing operations, to "unlock the UK's potential for augmented productivity", according to a new University of Cambridge report. Researchers argue that the UK currently lacks the computing capacity and capital required to build "generative" machine learning models fast enough to compete with US companies such as Google, Microsoft or Open AI. Instead, they call for a UK focus on leveraging these new AI systems for real-world applications – such as developing new diagnostic products and addressing the shortage of software engineers, for example – which could provide a major boost to the British economy. However, the researchers caution that without new legislation to ensure the UK has solid legal and ethical AI regulation, such plans could falter. British industries and the public may struggle to trust emerging AI platforms such as ChatGPT enough to invest time and money into skilling up. The policy report is a collaboration between Cambridge's Minderoo Centre for Technology and Democracy, Bennett Institute for Public Policy, and ai@cam: the University's flagship initiative on artificial intelligence.

With a low electrical hum, a small team of boxy, wheeled robots called "ants" criss-cross the top of a giant 3D grid of grey storage crates – 60,000 of them - ceaselessly arranging and rearranging them to order. Just one man, jokingly known as the robot whisperer, walks among them with a laptop. It would be hard to conceive of a more vivid example of robots taking on human jobs. "As robot technology advances, we can use them more and more, together with humans, to do useful work, and I think this is the future," says Jeroen Dekker, co-founder of Active Ants, the Dutch firm behind this newly opened e-commerce warehouse outside Northampton. "Yes, some jobs are disappearing, but that's the nasty jobs, for which we cannot find enough people."

Teleoperating a physical robot could become an important job in future, according to Sanctuary AI, based in Vancouver, Canada. The company also believes that this might provide a way to train robots how to perform tasks that are currently well out of their (mechanical) reach, and imbue machines with a physical sense of the world some argue is needed to unlock human-level artificial intelligence. Industrial robots are powerful, precise, and mostly stubbornly stupid. They cannot apply the kind of precision and responsiveness needed to perform delicate manipulation tasks. That's partly why the use of robots in factories is still relatively limited, and still requires an army of human workers to assemble all the fiddly bits into the guts of iPhones.

Knowledge graph embedding techniques are widely used for knowledge graph refinement tasks such as graph completion and triple classification. These techniques aim at embedding the entities and relations of a Knowledge Graph (KG) in a low dimensional continuous feature space. This paper adopts a transformer-based triplet network creating an embedding space that clusters the information about an entity or relation in the KG. It creates textual sequences from facts and fine-tunes a triplet network of pre-trained transformer-based language models. It adheres to an evaluation paradigm that relies on an efficient spatial semantic search technique. We show that this evaluation protocol is more adapted to a few-shot setting for the relation prediction task. Our proposed GilBERT method is evaluated on triplet classification and relation prediction tasks on multiple well-known benchmark knowledge graphs such as FB13, WN11, and FB15K. We show that GilBERT achieves better or comparable results to the state-of-the-art performance on these two refinement tasks.

Current deep learning models trained to generate radiology reports from chest radiographs are capable of producing clinically accurate, clear, and actionable text that can advance patient care. However, such systems all succumb to the same problem: making hallucinated references to non-existent prior reports. Such hallucinations occur because these models are trained on datasets of real-world patient reports that inherently refer to priors. To this end, we propose two methods to remove references to priors in radiology reports: (1) a GPT-3-based few-shot approach to rewrite medical reports without references to priors; and (2) a BioBERT-based token classification approach to directly remove words referring to priors. We use the aforementioned approaches to modify MIMIC-CXR, a publicly available dataset of chest X-rays and their associated free-text radiology reports; we then retrain CXR-RePaiR, a radiology report generation system, on the adapted MIMIC-CXR dataset. We find that our re-trained model--which we call CXR-ReDonE--outperforms previous report generation methods on clinical metrics, achieving an average BERTScore of 0.2351 (2.57% absolute improvement). We expect our approach to be broadly valuable in enabling current radiology report generation systems to be more directly integrated into clinical pipelines.

Transformers have recently been shown to generate high quality images from text input. However, the existing method of pose conditioning using skeleton image tokens is computationally inefficient and generate low quality images. Therefore we propose a new method; Keypoint Pose Encoding (KPE); KPE is 10 times more memory efficient and over 73% faster at generating high quality images from text input conditioned on the pose. The pose constraint improves the image quality and reduces errors on body extremities such as arms and legs. The additional benefits include invariance to changes in the target image domain and image resolution, making it easily scalable to higher resolution images. We demonstrate the versatility of KPE by generating photorealistic multiperson images derived from the DeepFashion dataset. We also introduce a evaluation method People Count Error (PCE) that is effective in detecting error in generated human images.