The future of artificial intelligence (AI) is both exciting and uncertain.

Audi claims to be reinventing the wheel designing using its in-house artificial intelligence technology FelGAN.

It even uses machine learning to highlight the most important parts of a terms of service agreement and … More from Artificial Intelligence.

Since OpenAI released ChatGPT last month, the text-based artificial intelligence tool has gone viral, attracting a million users in its first five …

Anyone who writes for a living, or simply enjoys good writing, has probably given some thought to the notion that artificial intelligence could …

The team behind GPT's new Chatbot has clearly done what they can to stop it spreading disinformation, but it is also quite clear at a level where we can say that that any nefarious commercial or governmental organization who wanted to weaponize these technologies for disinformation absolutely could. The first thing that GPT Chat demonstrates is a real confidence it its error, which is what we should expect from a machine. This is perfect troll behaviour. Not simply getting something wrong, but then (incorrectly) linking some elements to reinforce its point. Now with GPT Chat this is accidental, but it shows how you could easily bias the training data to support a proscribed position.

As marketers hurtle towards a privacy-first future, the industry is being flooded with countless stories about new approaches to consumer engagement that will soon shake up the advertising world. What if we need to go back to basics? What if it's time to return to contextual targeting? Marketing pros would understandably have some concerns. Contextual targeting has been around for a while and fell out of favor with many in the industry.

What the hell are we doing? All the best things are shutting down and people are out here using computers to make children's books? Not to get too Jeff Goldblum on you, but just because you build a tool to do something doesn't make it interesting or worthy. To wit, a product designer named Ammaar Reshi (previously employed by Peter Thiel at Palantir) has used AI to create a children's book in praise of AI… and it is fatuous and ugly. Unsurprisingly, those of us who actually like the human touch in art aren't thrilled.

In this paper, we present a study of regret and its expression on social media platforms. Specifically, we present a novel dataset of Reddit texts that have been classified into three classes: Regret by Action, Regret by Inaction, and No Regret. We then use this dataset to investigate the language used to express regret on Reddit and to identify the domains of text that are most commonly associated with regret. Our findings show that Reddit users are most likely to express regret for past actions, particularly in the domain of relationships. We also found that deep learning models using GloVe embedding outperformed other models in all experiments, indicating the effectiveness of GloVe for representing the meaning and context of words in the domain of regret. Overall, our study provides valuable insights into the nature and prevalence of regret on social media, as well as the potential of deep learning and word embeddings for analyzing and understanding emotional language in online text. These findings have implications for the development of natural language processing algorithms and the design of social media platforms that support emotional expression and communication.

We propose a fully unsupervised method to detect bias in contextualized embeddings. The method leverages the assortative information latently encoded by social networks and combines orthogonality regularization, structured sparsity learning, and graph neural networks to find the embedding subspace capturing this information. As a concrete example, we focus on the phenomenon of ideological bias: we introduce the concept of an ideological subspace, show how it can be found by applying our method to online discussion forums, and present techniques to probe it. Our experiments suggest that the ideological subspace encodes abstract evaluative semantics and reflects changes in the political left-right spectrum during the presidency of Donald Trump.