OpenAI's messy firing and re-hiring of its powerful chief executive this week shocked the tech world. But the power struggle has implications beyond the company's boardroom, AI experts said. It throws into relief the greenness of the AI industry and the strong desire in Silicon Valley to be first, and raises urgent questions about the safety of the technology. "The AI that we're looking at now is immature. There are no standards, no professional body, no certifications. Everybody figures out how to do it, figures out their own internal norms," said Rayid Ghani, a professor of machine learning and public policy at Carnegie Mellon University.

OpenAI was reportedly working on an advanced system before Sam Altman's sacking that was so powerful it caused safety concerns among staff at the company. The artificial intelligence model triggered such alarm with some OpenAI researchers that they wrote to the board of directors before Altman's dismissal warning it could threaten humanity, Reuters reported. The model, called Q* – and pronounced as "Q-Star" – was able to solve basic maths problems it had not seen before, according to the tech news site the Information, which added that the pace of development behind the system had alarmed some safety researchers. The ability to solve maths problems would be viewed as a significant development in AI. The reports followed days of turmoil at San Francisco-based OpenAI, whose board sacked Altman last Friday but then reinstated him on Tuesday night after nearly all the company's 750 staff threatened to resign if he was not brought back.

The American tech industry has long paid reverence to its monolithic slate of founders and visionaries: Meta's Mark Zuckerberg; Google's Larry Page and Sergey Brin; Apple's Steve Jobs and Tim Cook. But where the other firms sold phones and search engines, Musk and Altman championed their work as a public mission for protecting mankind, with a for-profit business attached. It is notable that as private companies, they don't have to report to federal regulators or to shareholders, who can vote down proposals or push back against their work.

Sam Altman is back at the helm of OpenAI, days after the board abruptly ousted him. Almost everything else is still in flux. The deal struck Tuesday night to restore Altman as CEO is a long way from the ultimate goal he wanted to achieve heading into a weekend of intense negotiations. He had lobbied for an entirely new slate of directors--built on the ashes of the directors who fired him--and wanted to rejoin the board again himself, according to people familiar with the matter.

This week, Emily Bazelon, John Dickerson, and David Plotz discuss the problems with issue polling and issues with political journalism; the chaos and conflict of Sam Altman and OpenAI; and the failure of the Oslo Accords and perpetual struggle between Israel and Palestine. Send us your Conundrums: submit them at slate.com/conundrum. And join us in-person or online with our special guest – The Late Show's Steven Colbert – for Gabfest Live: The Conundrums Edition! December 7 at The 92nd Street Y, New York City. Here are some notes and references from this week's show: Nate Cohn for The New York Times: The Crisis in Issue Polling, and What We're Doing About It and We Did an Experiment to See How Much Democracy and Abortion Matter to Voters Eli Saslow for The New York Times: A Jan. 6 Defendant Pleads His Case to the Son Who Turned Him In John Dickerson and Jo Ling Kent for CBS News Prime Time: What Sam Altman's ouster from OpenAI could mean for the tech world Emily Bazelon for The New York Times Magazine: Was Peace Ever Possible? Ezra Klein for The New York Times's The Ezra Klein Show podcast: The Best Primer I've Heard on Israeli-Palestinian Peace Efforts John Dickerson for CBS Mornings: Former President Jimmy Carter: "America will learn from its mistakes" Here are this week's chatters: John: Julia Simon for NPR: 'It feels like I'm not crazy.'

Ahead of OpenAI CEO Sam Altman's four days in exile, several staff researchers wrote a letter to the board of directors warning of a powerful artificial intelligence discovery that they said could threaten humanity, two people familiar with the matter said. The previously unreported letter and AI algorithm were key developments before the board's ouster of Altman, the poster child of generative AI, the two sources said. Prior to his triumphant return late Tuesday, more than 700 employees had threatened to quit and join backer Microsoft in solidarity with their fired leader. The sources cited the letter as one factor among a longer list of grievances by the board leading to Altman's firing, among which were concerns over commercializing advances before understanding the consequences. A copy of the letter was unable to be reviewed for this report.

Sam Altman is returning to lead OpenAI less than five days after his surprise dismissal, which kicked off a tug of war for his talent, left the company in disarray and laid bare deep board divisions over the mission of one of the world's most valuable startups. OpenAI's new interim board, which won't include Altman at the outset, will be led by Bret Taylor, a former co-CEO of Salesforce. The other directors are Larry Summers, the former U.S. treasury secretary, and existing member Adam D'Angelo, the co-founder and CEO of Quora. Altman had been fired Friday after clashing with the board over his drive to transform OpenAI from a nonprofit organization focused on the scientific exploration of artificial intelligence into a business that builds products, attracts customers and lines up the funding needed to power AI tools. Members of the former board harbored concerns about the potential harms done by powerful, unchecked AI.

Optimal Transport has received much attention in Machine Learning as it allows to compare probability distributions by exploiting the geometry of the underlying space. However, in its original formulation, solving this problem suffers from a significant computational burden. Thus, a meaningful line of work consists at proposing alternatives to reduce this burden while still enjoying its properties. In this thesis, we focus on alternatives which use projections on subspaces. The main such alternative is the Sliced-Wasserstein distance, which we first propose to extend to Riemannian manifolds in order to use it in Machine Learning applications for which using such spaces has been shown to be beneficial in the recent years. We also study sliced distances between positive measures in the so-called unbalanced OT problem. Back to the original Euclidean Sliced-Wasserstein distance between probability measures, we study the dynamic of gradient flows when endowing the space with this distance in place of the usual Wasserstein distance. Then, we investigate the use of the Busemann function, a generalization of the inner product in metric spaces, in the space of probability measures. Finally, we extend the subspace detour approach to incomparable spaces using the Gromov-Wasserstein distance.

Task-oriented conversational datasets often lack topic variability and linguistic diversity. However, with the advent of Large Language Models (LLMs) pretrained on extensive, multilingual and diverse text data, these limitations seem overcome. Nevertheless, their generalisability to different languages and domains in dialogue applications remains uncertain without benchmarking datasets. This paper presents a holistic annotation approach for emotion and conversational quality in the context of bilingual customer support conversations. By performing annotations that take into consideration the complete instances that compose a conversation, one can form a broader perspective of the dialogue as a whole. Furthermore, it provides a unique and valuable resource for the development of text classification models. To this end, we present benchmarks for Emotion Recognition and Dialogue Quality Estimation and show that further research is needed to leverage these models in a production setting.

Generating high-quality summaries for chat dialogs often requires large labeled datasets. We propose a method to efficiently use unlabeled data for extractive summarization of customer-agent dialogs. In our method, we frame summarization as a question-answering problem and use state-of-the-art large language models (LLMs) to generate pseudo-labels for a dialog. We then use these pseudo-labels to fine-tune a chat summarization model, effectively transferring knowledge from the large LLM into a smaller specialized model. We demonstrate our method on the \tweetsumm dataset, and show that using 10% of the original labelled data set we can achieve 65.9/57.0/61.0 ROUGE-1/-2/-L, whereas the current state-of-the-art trained on the entire training data set obtains 65.16/55.81/64.37 ROUGE-1/-2/-L. In other words, in the worst case (i.e., ROUGE-L) we still effectively retain 94.7% of the performance while using only 10% of the data.