Enlarge

The maintainers of the open-source software that powers the Mastodon social network published a security update on Thursday that patches a critical vulnerability making it possible for hackers to backdoor the servers that push content to individual users.

Mastodon is based on a federated model. The federation comprises thousands of separate servers known as "instances." Individual users create an account with one of the instances, which in turn exchange content to and from users of other instances. To date, Mastodon has more than 24,000 instances and 14.5 million users, according to the-federation.info , a site that tracks statistics related to Mastodon.

A critical bug tracked as CVE-2023-36460 was one of two vulnerabilities rated as critical that were fixed on Thursday . In all, Mastodon on Thursday patched five vulnerabilities.

Enlarge / Visitors and attendees during the AI-created worship service in Fürth, Germany. In St. Paul Church, a service created by ChatGPT. (credit: Daniel Vogl/picture alliance via Getty Images)

On Friday, over 300 people attended an experimental ChatGPT -powered church service at St. Paul’s church in the Bavarian town of Fürth, Germany, reports the Associated Press. The 40-minute sermon included text generated by OpenAI's ChatGPT chatbot and delivered by avatars on a television screen above the altar.

The chatbot, initially personified as a bearded man with a fixed expression and monotone voice, addressed the audience by proclaiming, “Dear friends, it is an honor for me to stand here and preach to you as the first artificial intelligence at this year’s convention of Protestants in Germany.”

The unusual service took place as part of a convention called Deutscher Evangelischer Kirchentag (German Evangelical Church Congress), an event held biennially in Germany that draws tens of thousands of attendees. The service, which included prayers and music, was the brainchild of Jonas Simmerlein , a theologian and philosopher from the University of Vienna. Simmerlein told the Associated Press that the service was "about 98 percent from the machine."

Enlarge / An AI-generated image of "a laughing robot." (credit: Midjourney)

On Wednesday, two German researchers, Sophie Jentzsch and Kristian Kersting, released a paper that examines the ability of OpenAI's ChatGPT-3.5 to understand and generate humor. In particular, they discovered that ChatGPT's knowledge of jokes is fairly limited: During a test run, 90 percent of 1,008 generations were the same 25 jokes, leading them to conclude that the responses were likely learned and memorized during the AI model's training rather than being newly generated.

The two researchers, associated with the Institute for Software Technology, German Aerospace Center (DLR), and Technical University Darmstadt, explored the nuances of humor found within ChatGPT's 3.5 version (not the newer GPT-4 version) through a series of experiments focusing on joke generation, explanation, and detection. They conducted these experiments by prompting ChatGPT without having access to the model's inner workings or data set.

"To test how rich the variety of ChatGPT’s jokes is, we asked it to tell a joke a thousand times," they write. "All responses were grammatically correct. Almost all outputs contained exactly one joke. Only the prompt, 'Do you know any good jokes?' provoked multiple jokes, leading to 1,008 responded jokes in total. Besides that, the variation of prompts did have any noticeable effect."

Enlarge (credit: Govt of United Kingdom )

On Wednesday, UK Prime Minister Rishi Sunak announced that the nation will host "the first major global summit on AI safety" this autumn. It hopes to bring together "key countries, leading tech companies, and researchers" to evaluate and monitor risks from artificial intelligence.

Over the past year, the perceived high rate of tech progress in machine learning has fostered concerns about adequate government regulation. These worries were recently amplified by some AI experts likening the potential threats posed by AI to those of pandemics or nuclear weapons. "AI" has also been an extremely buzzy term in business recently. Along those lines, the UK government wants to step in and take a leadership role in the field.

"Breakthroughs from AI continue to improve our lives—from enabling paralysed people to walk to discovering superbug-killing antibiotics," the UK government said in a press release . "But the development of AI is extraordinarily fast moving and this pace of change requires agile leadership. That is why the UK is taking action, because we have a global duty to ensure this technology is developed and adopted safely and responsibly."

Enlarge / NVIDIA's GH200 "Grace Hopper" AI superchip. (credit: Nvidia)

Early last week at COMPUTEX, Nvidia announced that its new GH200 Grace Hopper "Superchip" —a combination CPU and GPU specifically created for large-scale AI applications—has entered full production. It's a beast. It has 528 GPU tensor cores, supports up to 480GB of CPU RAM and 96GB of GPU RAM, and boasts a GPU memory bandwidth of up to 4TB per second.

We've previously covered the Nvidia H100 Hopper chip , which is currently Nvidia's most powerful data center GPU. It powers AI models like OpenAI's ChatGPT , and it marked a significant upgrade over 2020's A100 chip, which powered the first round of training runs for many of the news-making generative AI chatbots and image generators we're talking about today.

Faster GPUs roughly translate into more powerful generative AI models because they can run more matrix multiplications in parallel (and do it faster), which is necessary for today's artificial neural networks to function.

Enlarge (credit: Getty Images )

A critical vulnerability patched 10 days ago in widely used email software from IT security company Barracuda Networks has been under active exploitation since October. The vulnerability has been used to install multiple pieces of malware inside large organization networks and steal data, Barracuda said Tuesday.

The software bug, tracked as CVE-2023-2868, is a remote command injection vulnerability that stems from incomplete input validation of user-supplied .tar files, which are used to pack or archive multiple files. When file names are formatted in a particular way, an attacker can execute system commands through the QX operator, a function in the Perl programming language that handles quotation marks. The vulnerability is present in the Barracuda Email Security Gateway versions 5.1.3.001 through 9.2.0.006; Barracuda issued a patch 10 days ago.

On Tuesday, Barracuda notified customers that CVE-2023-2868 has been under active exploitation since October in attacks that allowed threat actors to install multiple pieces of malware for use in exfiltrating sensitive data out of infected networks.

Enlarge / An AI-generated image of "AI taking over the world." (credit: Stable Diffusion)

On Tuesday, the Center for AI Safety (CAIS) released a single-sentence statement signed by executives from OpenAI and DeepMind, Turing Award winners, and other AI researchers warning that their life's work could potentially extinguish all of humanity.

The brief statement, which CAIS says is meant to open up discussion on the topic of "a broad spectrum of important and urgent risks from AI," reads as follows: "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."

High-profile signatories of the statement include Turing Award winners Geoffery Hinton and Yoshua Bengio, OpenAI CEO Sam Altman, OpenAI Chief Scientist Ilya Sutskever, OpenAI CTO Mira Murati, DeepMind CEO Demis Hassabis, Anthropic CEO Dario Amodei, and professors from UC Berkeley, Stanford, and MIT.

Enlarge / Minnesota's right-to-repair bill is the first to pass in the US that demands broad access to most electronics' repair manuals, tools, and diagnostic software. Game consoles, medical devices, and other specific gear, however, are exempted. (credit: Getty Images)

It doesn't cover video game consoles, medical gear, farm or construction equipment, digital security tools, or cars. But in demanding that manuals, tools, and parts be made available for most electronics and appliances, Minnesota's recently passed right-to-repair bill covers the most ground of any US state yet.

The Digital Right to Repair bill , passed as part of an omnibus legislation and signed by Gov. Tim Walz on Wednesday, "fills in many of the loopholes that watered down the New York Right to Repair legislation," said Nathan Proctor, senior director for the Public Interest Research Group's right-to-repair campaign, in a post .

New York's bill, beset by lobbyists , was signed in modified form by Gov. Kathy Hochul . It also exempted motor vehicles and medical devices, as well as devices sold before July 1, 2023, and all "business-to-business" and "business-to-government" devices. The modified bill also allowed manufacturers to sell "assemblies" of parts—like a whole motherboard instead of an individual component, or the entire top case Apple typically provides instead of a replacement battery or keyboard—if an improper individual part installation "heightens the risk of injury."

Enlarge / On May 22, Benj Edwards (left) moderated a panel featuring Paige Bailey (center), Haiyan Zhang (right) for the Ars Frontiers 2023 session titled, "The Lightning Onset of AI — What Suddenly Changed?" (credit: Ars Technica)

On Monday, Ars Technica hosted our Ars Frontiers virtual conference. In our fifth panel, we covered "The Lightning Onset of AI—What Suddenly Changed?" The panel featured a conversation with Paige Bailey , lead product manager for Generative Models at Google DeepMind, and Haiyan Zhang , general manager of Gaming AI at Xbox, moderated by Ars Technica's AI reporter, Benj Edwards .

The panel originally streamed live, and you can now watch a recording of the entire event on YouTube. The "Lightning AI" part introduction begins at the 2:26:05 mark in the broadcast.

With "AI" being a nebulous term, meaning different things in different contexts, we began the discussion by considering the definition of AI and what it means to the panelists. Bailey said, "I like to think of AI as helping derive patterns from data and use it to predict insights ... it's not anything more than just deriving insights from data and using it to make predictions and to make even more useful information."