Goto

Collaborating Authors

 Government


Predicting Hateful Discussions on Reddit using Graph Transformer Networks and Communal Context

arXiv.org Artificial Intelligence

We propose a system to predict harmful discussions on social media platforms. Our solution uses contextual deep language models and proposes the novel idea of integrating state-of-the-art Graph Transformer Networks to analyze all conversations that follow an initial post. This framework also supports adapting to future comments as the conversation unfolds. In addition, we study whether a community-specific analysis of hate speech leads to more effective detection of hateful discussions. We evaluate our approach on 333,487 Reddit discussions from various communities. We find that community-specific modeling improves performance two-fold and that models which capture wider-discussion context improve accuracy by 28\% (35\% for the most hateful content) compared to limited context models.


AI experts see legislation moving forward in 2023

#artificialintelligence

Various legislators and policy experts have recently shared their views on the imminent future of artificial intelligence (AI) legislation. Speaking at CES 2023, as quoted by IoT World Today, Laura Caroli, a parliamentary assistant currently leading negotiations on the AI Act, says an agreement is expected by March. Regarding biometrics, the Council of the EU tweaked the remote biometric identification system definition in December 2022 by clarifying that such systems can be used only in cases where it is "strictly necessary for law enforcement purposes." Caroli also adds that while the legislation has now been approved by the Council of the European Union and should be fully approved by the end of 2023, the law will only come into force two years later. "It takes a long time, but it's a complex piece of legislation and a lot of tension politically."


The weirdest news of 2022 - Cloud7 News

#artificialintelligence

In 2022, we saw many new technologies and methods in the tech industry and in the never-ending battle between hackers and cybersecurity experts. Sometimes, these new technologies can result in unexpected outcomes, or discovering mind-blowing methods seems like a scene from science fiction movies. Our lists include a music video becoming a vulnerability with a CVE, very unusual services being hacked by hackers, an AI bot turning into a hacker's assistant, and many more. Each year, ransomware groups are attacks thousands of organizations to extort money. But this time, the perpetrator became the victim.


DoNotPay Offers Lawyers $1M to Let Its AI Argue Before Supreme Court

#artificialintelligence

A legal services company says it's willing to pay $1 million to fuck around and find out. On Sunday, DoNotPay CEO Joshua Browder made a wild proposition to any lawyer slated to argue an upcoming case in front of the U.S. Supreme Court. Let DoNotPay's AI lawyer, which is built on OpenAI's viral GPT-3 API, argue the case before the court, Browder said, in exchange for $1 million. All the human lawyer would need to do is wear AirPods and repeat to the court what DoNotPay's robot lawyer argues. "DoNotPay will pay any lawyer or person $1,000,000 with an upcoming case in front of the United States Supreme Court to wear AirPods and let our robot lawyer argue the case by repeating exactly what it says," Browder wrote on Twitter on Sunday night.


Meta rolls out AI ad-targeting tech in an effort to reduce discrimination

Engadget

Meta is acting on its vow to reduce ad discrimination through technology. The company is rolling out a Variance Reduction System (VRS) in the US that ensures the real audience for an ad more closely matches the eligible target audience -- that is, it shouldn't skew unfairly toward certain cultural groups. Once enough people have seen an ad, a machine learning system compares the aggregate demographics of viewers with those the marketers intended to reach. It then tweaks the ad's auction value (that is, the likelihood you'll see the ad) to display it more or less often to certain groups. VRS keeps working throughout an ad run.


Russia-Ukraine War an outlook for 2023: more bloodshed to come with no end in sight

FOX News

Former Defense Intelligence Agency Officer Rebekah Koffler discusses why peace talks are unlikely between Russia, Ukraine, and the U.S., on'Varney & Co.' As we have welcome the New Year, many on both sides of the Atlantic are wondering whether the Russia-Ukraine conflict, the biggest war in Europe since World War II, will come to an end this year. The largest country on the continent, Ukraine, is being depopulated, having lost more than 100,000 of its citizens to death or injury. Europe itself is being destabilized by financial woes and influx of refugees from war-torn areas. Contrary to the hopes of many, not only will 2023 not bring peace, it will likely see the most bloodshed yet, as the key warring parties – Moscow, Kyiv, and Washington, D.C. – are all postured for decisive escalation. Here's why we are probably entering the "hottest" phase of this war.


The Download: introducing our 10 Breakthrough Technologies

MIT Technology Review

Each year, MIT Technology Review's reporters and editors pick 10 Breakthrough Technologies, all of which have the promise to fundamentally change the way we live and work. This year's list covers everything from space science and telemedicine to advances in artificial intelligence and biotechnology. They represent the technologies we predict will have the biggest impact on our lives in the year ahead. This year's TR10 is the 22nd we've published, and I'll be highlighting an entry each day in The Download for the next 10 days, starting tomorrow. We hope you enjoy marveling at the progress that's been made in the fields of gene editing, military drones, battery recycling, and computer chip design, to name just a few. David Rotman, our editor at large, has written a fascinating introductory essay which sets out how legislation investing hundreds of billions into industry and research and development could reset how we think about governments' role in the economy.


The Morning After: The best of CES 2023

Engadget

The top news stories today: Apple’s long-rumored mixed reality headset could finally debut this spring, Samsung might unveil the Galaxy S23 series on February 1st, The best of CES 2023..


Four types of bias in medical AI are running under the FDA's radar – STAT News

#artificialintelligence

Although artificial intelligence is entering health care with great … In technical terms, what we're describing is called a machine learning …


Activ Surgical completes first AI-enabled case with ActivSight intelligent light

#artificialintelligence

Boston-based Activ Surgical designed ActivSight to provide enhanced visualization and real-time on-demand surgical insights in the operating room. The easy-to-adapt model seamlessly attaches to laparoscopic and robotic systems. Activ Surgical received CE mark approval for ActivSight in December 2022. The FDA cleared the technology in April 2021. The company aims to transform the surgical experience using emerging technologies and data.