A company behind an AI-powered app called CrimeRadar has apologised for the distress caused by false crime alerts issued to local US communities after a BBC Verify investigation. CrimeRadar uses artificial intelligence to monitor openly available police radio communications, automatically generating a transcript and then producing crime alerts for users across the US. BBC Verify has found multiple instances from Florida to Oregon of CrimeRadar sending misleading and inaccurate alerts about serious crime to local residents - as Thomas Copeland explains. The barge that wrecked in 1918, famous for a dramatic rescue, is now shifting closer to the falls as recent movements carry it further from its original resting spot. House Speaker Mike Johnson, Hakeem Jeffries and Chuck Schumer participated in the annual ceremony that commemorates the eight days of Hanukkah.

Adm. Daryl Caudle joins'America's Newsroom' to discuss rising tensions with China's navy, the use of AI in US defense, and a powerful Memorial Day re-enlistment ceremony at Ground Zero. The U.S. Navy has called off a search for a sailor assigned to the USS George Washington amid reports that he possibly went overboard while the ship was sailing north of Australia. The sailor was reported overboard on the aircraft carrier on Monday as the ship was transiting the Timor Sea, the Navy said. US DEFENSE OFFICIAL REACTS TO IRAN'S CLAIMS ABOUT ENCOUNTER WITH WARSHIP This photo shows a general view of U.S. aircraft carrier USS George Washington shortly after berthing at Manila Bay in Manila on July 3. (TED ALJIBE/AFP via Getty Images) The search effort involving the George Washington, its carrier strike group, as well as the Australian Defence (sic) Force and Australian Border Force, concluded at 12:40 p.m. Wednesday. "USS George Washington expresses sincere condolences to those impacted by this loss and is actively engaged with the crew to make services available to tend to their needs during this challenging time," Lt. Cmdr.

On February 19th, Donald Trump logged onto Truth Social to congratulate himself on vanquishing congestion pricing in his home state. "CONGESTION PRICING IS DEAD," he posted. "Manhattan, and all of New York, is SAVED. The message was amplified by the White House's official X account, which tweeted it with an A.I.-generated image of Trump, golden-haired and golden-crowned, blotting out the New York City skyline. The illustration, which was styled to look like the cover of Time magazine, displayed the President's fondness for crude symbols of power and wealth.

President-elect Donald Trump touted his recent Cabinet picks as he prepares his White House return. The Fox & Friends co-hosts react. Sylvester Stallone is warning his fans after a fake trailer for "The Godfather Part 4" went viral online. Stallone, 78, took to social media to comment on the fan-made video creation. Lol this is definitely not to be taken seriously!" the Hollywood actor laughed and wrote on Instagram. His post included two photos of Stallone, one of him smoking a cigar and the second of the actor holding a gun. Sylvester Stallone sent a message to his fans after a fake trailer for "The Godfather Part 4" went viral online. The creator of the fake "Godfather 4" video trailer sent a message to viewers explaining how it was made. "Please note that this video is a concept trailer created solely for artistic and entertainment purposes.

Retrieval-augmented generation (RAG) methods are viable solutions for addressing the static memory limits of pre-trained language models. Nevertheless, encountering conflicting sources of information within the retrieval context is an inevitable practical challenge. In such situations, the language models are recommended to transparently inform users about the conflicts rather than autonomously deciding what to present based on their inherent biases. To analyze how current large language models (LLMs) align with our recommendation, we introduce WhoQA, a public benchmark dataset to examine model's behavior in knowledge conflict situations. We induce conflicts by asking about a common property among entities having the same name, resulting in questions with up to 8 distinctive answers. WhoQA evaluation set includes 5K questions across 13 Wikidata property types and 150K Wikipedia entities. Our experiments show that despite the simplicity of WhoQA questions, knowledge conflicts significantly degrades LLMs' performance in RAG settings.

Rep. Alexandria Ocasio-Cortez, D-N.Y., emphasized the possibility that Vice President Kamala Harris will be the "first woman President of the United States" during a late-night appearance with Stephen Colbert. Rep. Alexandria Ocasio-Cortez, D-N.Y., predicted that Vice President Kamala Harris will be the "first woman President of the United States" during a late-night appearance on Thursday, conspicuously leaning into identity politics. Following the end of the Democratic National Convention, the progressive lawmaker went on CBS' "The Late Show with Stephen Colbert" and played up how remarkable it is that Harris could become president. We will have the first woman President of the United States in November," Ocasio-Cortez predicted to raucous applause. Rep. Alexandria Ocasio-Cortez, D-N.Y., predicts that Vice President Kamala Harris will be the first woman president. Playing up the drama in her declaration, the congresswoman talked about how she grew up only seeing depictions of female leadership in episodes of "Star Trek: Voyager" that she watched with her dad as a kid. But now that Harris has been nominated, that fantasy is one step closer to reality. "My dad felt it very important for me to watch this because he wanted me to see an example of a woman in leadership, and when I was a kid the only example of that was in science fiction." She continued, "And today represents a day where it has become our reality." The late-night show audience went wild at the prospect. She also had a welcome interviewer in Colbert, a rabid Democratic supporter who even moderated a fundraiser for President Biden before he was forced off the 2024 ticket. Colbert then asked the lawmaker about her statements earlier this year warning that the Democratic Party will not unite behind Harris if President Biden steps aside. Reading her quote, he said, "'If you think that there is a consensus among the people who want Joe Biden to leave that they will support Vice President Harris, you will be mistaken." The host then asked her, "Have you ever been happier to be wrong?" She replied, repeating the word, "Ecstatic" several times. CNN'S DANA BASH ARGUES DNC APPEALS TO MEN WHO ARE NOT SO'TESTOSTERONE-LADEN' Colbert continued, "You didn't think this would necessarily happen?" to which she said, "No, and I think it is important that this was not predestined, this was not predetermined." "Vice President Harris earned this nomination through her grit, her politics, through every bit of hard work.

GPT-3, the highly-touted text generator built by OpenAI, can do a lot of things. For example, Microsoft today announced a new AI-powered "autocomplete" system for coding that uses GPT-3 to build out code solutions for people without requiring them to do any developing. But one thing the technology can not do is "dupe humans" with its ability to write misinformation. Yet, you wouldn't know that if you were solely judging by the headlines in your news feed. Wired recently ran an article with the title "GPT-3 can write disinformation now – and dupe human readers," and it was picked up by other outlets who then reflected the coverage.

Most people assume that AI is dispassionate, unbiased, objective rather than subjective, and indubitably a truth-teller. Often fostered by those sci-fi AI-based robots portrayed in movies, it seems that AI itself has managed to gain an assuredly unearned reputation, perhaps promulgated by a terrific Hollywood agent that keeps AI in generally good stead with the public-at-large (well, other than the cases where the AI decides to wipe out humanity and expunge us from the earth, a rather hurtful act, one would so suppose). In the real world, there is no particular reason to believe that AI is going to be telling the truth. Anyone that interacts with an AI system can get themselves into quite an untoward posture and suffer inadvertent adverse consequences by doting on every utterance of an AI system and blithely accepting such interaction as some form of absolute truth. Let's start with the now known facet about AI potentially containing biases and make our way to the unnerving aspect about AI lying through-its-teeth.

Multilingual pre-trained models could leverage the training data from a rich source language (such as English) to improve performance on low resource languages. However, the transfer quality for multilingual Machine Reading Comprehension (MRC) is significantly worse than sentence classification tasks mainly due to the requirement of MRC to detect the word level answer boundary. In this paper, we propose two auxiliary tasks in the fine-tuning stage to create additional phrase boundary supervision: (1) A mixed MRC task, which translates the question or passage to other languages and builds cross-lingual question-passage pairs; (2) A language-agnostic knowledge masking task by leveraging knowledge phrases mined from web. Besides, extensive experiments on two cross-lingual MRC datasets show the effectiveness of our proposed approach.

GARY MARCUS: The dominant vision in the field right now is, collect a lot of data, run a lot of statistics, and intelligence will emerge. And I think that's wrong. I think that having a lot of data is important, and collecting a lot of statistics is important. But I think what we also need is deep understanding, not just so-called "deep learning." So deep learning finds what's typically correlated, but we all know that correlation is not the same thing as causation.