Saturday, July 8, 2023

OpenAI Social Media Horse Races Biden Administration More: Saturday Afternoon ResearchBuzz July 8 2023

OpenAI, Social Media Horse Races, Biden Administration, More: Saturday Afternoon ResearchBuzz, July 8, 2023
By ResearchBuzz

TWEAKS AND UPDATES

TechCrunch: OpenAI makes GPT-4 generally available. “OpenAI today announced the general availability of GPT-4, its latest text-generating model, through its API. Starting this afternoon, all existing OpenAI API developers ‘with a history of successful payments’ can access GPT-4. The company plans to open up access to new developers by the end of this month, and then start raising availability limits after that ‘depending on compute availability.'”

Daily Beast: Twitter Star Journos ‘Cautiously’ Move Over to Threads. “While Twitter-aping revivals including Bluesky, Mastodon, Post News, Hive Social, and Substack Notes have sputtered out of relevance, Threads has garnered an astonishing 48 million sign-ups within 24 hours, according to The Verge. Its rise has even prompted Twitter to send a cease-and-desist letter to the company, arguing Threads is predicated on Twitter’s trade secrets.”

Associated Press: Biden administration seeks stay of judge’s social media order, saying it could cause ‘grave harm’. “A Louisiana-based federal judge’s order broadly limiting executive branch communications with social media companies could cause ‘grave harm’ by preventing the government from ‘engaging in a vast range of lawful and responsible conduct,’ Biden administration attorneys said in a motion filed Thursday with a federal appeals court.”

AROUND THE INTERNET WORLD

The Guardian: New Twitter rules restrict US weather service, raising safety fears. “Twitter’s new volume limits on viewing posts suddenly left several National Weather Service (NWS) offices across the US unable to receive tweets from storm spotters who help with tracking extreme weather, including during storms this week – prompting safety warnings.”

Los Angeles Times: Santa Monica’s Headspace Health laid off scores of therapists. Their patients don’t know where they went. “When Headspace Health laid off a large number of its therapists June 29, patients were told their providers had left the platform. What they didn’t know was their therapists had lost their jobs. And they suddenly had no way to contact them.”

Trinity College Dublin: Unlocking the Fagel Collection – Trinity’s Old Library celebrates its Dutch treasures. “Botanical catalogues, lavish celestial atlases and unique pamphlets from the early modern period are among 30,000 titles being conserved and digitally catalogued in an ambitious collaboration to register the entirety of the 18th-century Fagel Collection, which fills a mile of shelving space in the Old Library of Trinity College Dublin.”

SECURITY & LEGAL

The Hill: FBI launches national ‘swatting’ database amid rising incidents. “The FBI is tracking ‘swatting’ incidents in a national database as the dangerous form of prank call becomes more common, the bureau revealed Thursday. Swatting incidents take place when a person calls the police claiming there is a dangerous person, kidnapping or a mass shooting at a house, hoping for police to respond in force.”

CBS News: “Mom influencer” Katie Sorensen sentenced to jail for falsely claiming couple tried to kidnap her kids at a crafts store. “A social media influencer who went viral for a 2020 Instagram video falsely claiming a couple tried to kidnap her children at a crafts store in Northern California has been sentenced to jail… Kathleen ‘Katie’ Sorensen, 30, was sentenced to 90 days in jail, 60 of which could be served on a work-release program, the Sonoma County district attorney’s office said.”

The Verge: Telegram has become a window into war. “The messaging app has become a key channel for news about the Russian invasion of Ukraine. But the government’s relationship with it is complicated.”

RESEARCH & OPINION

MIT News: Researchers teach an AI to write better chart captions. “The MIT researchers found that machine-learning models trained for autocaptioning with their dataset consistently generated captions that were precise, semantically rich, and described data trends and complex patterns. Quantitative and qualitative analyses revealed that their models captioned charts more effectively than other autocaptioning systems.” Good afternoon, Internet…

Do you like ResearchBuzz? Does it help you out? Please consider supporting it on Patreon. Not interested in commitment? Perhaps you’d buy me an iced tea. Check out Search Gizmos when you have a minute.



July 9, 2023 at 12:10AM
via ResearchBuzz https://ift.tt/DrF6OxM

No comments:

Post a Comment