Artificial intelligence is getting everyone excited. It's going to end or improve the world, depending on your optimism/pessimism. The latest hullabaloo was triggered by the release of ChatGPT – the progression of so called generative AI, which doesn't just analyse data but actually creates new content (in this case written text). There's been lots of speculation of what this might mean for education (the end of coursework?), but my focus is on the implications for the labour market. Now the first serious research on that front has arrived.

On February 6, 2023, earthquakes measuring 7.7 and 7.6 hit South Eastern Turkey, affecting 10 cities and resulting in more than 42,000 deaths and 120,000 injured as of February 21. A few hours after the earthquake, a group of programmers started a Discord server to roll out an application called afetharita, literally meaning, disaster map. This application would serve search & rescue teams and volunteers to find survivors and bring them help. The need for such an app arose when survivors posted screenshots of texts with their addresses and what they needed (including rescue) on social media. Some survivors also tweeted what they needed so their relatives knew they were alive and that they need rescue.

ChatGPT feels human and is responding more and more like we do everyday. It can act like a friend, a therapist, and an advisor. We can have conversations with it and have it impersonate someone else so well we won't tell the difference. We may even ask "What would Steve Jobs say?" and get a reasonable response. Friends and family we can't contact or who passed away. There's someone in my life who isn't here and I'd do anything to have a conversation with them.

In the previous post, I explored how useful ChatGPT currently is for data visualisation. This was done through testing this AI tool on a number of tasks and on its information retrieval ability. The results showed that ChatGPT still had some way to go on the theory and consulting side of things, as it still made mistakes and had limited knowledge on the subject. Despite this, it's still impressive how it can easily provide a written response that was mostly correct for what I had asked. Where ChatGPT really shined was on the coding side of DataViz.

There has been significant buzz about ChatGPT over the past couple of months and quite remarkably so considering it was only launched on 30 November 2022. Perhaps of great significance to ChatGPT's meteoric rise has been the fact that it is open for public usage free of charge. This is mainly because the service is still in its research phase, and although that may seem attractive for users, it also means that ChatGPT is collecting a significant amount of data and personal information. ChatGPT is enabled by its language model. A language model is a probability distribution over a sequence of words.

Companies from Meta to Home Depot are flooding earnings calls with commentary about their artificial intelligence efforts. Ever since OpenAI's ChatGPT lit up the internet in November, companies can't stop talking about artificial intelligence. Take this earnings season so far: References to AI and related terms during calls with investors are already up 77% from a year earlier. AI-hungry investors have propelled Nvidia Corp., which makes the chips needed for complex AI computing tasks, into the best-performing stock among mega-caps this year. Relatively obscure firms with AI in their names have also skyrocketed.

ChatGPT and similar systems will increasingly be part of our lives, including health care. We need guidelines to ensure their ethical deployment. Generative AI systems like ChatGPT, a chatbot based on a generative pre-trained transformer, have captured the public's attention, resulting in a flurry of positive and negative speculation about their potential. They have even found their way into popular comic strips. In one Dilbert strip, for instance, the boss asks Wally if his status report was written by a commercial grade AI.

I've decided to step back from the OpenAI board due to the potential for conflicts of interest with my role as an investor with Greylock. I remain an ally to OpenAI and its mission of beneficial AI for humanity. AI, like most transformative technologies, grows gradually, then arrives suddenly. Headlines make AI feel abrupt and singular when it's compared to a tidal… 35 comments on LinkedIn

Large language models like ChatGPT are purveyors of plausibility. The chatbots, many based on so-called generative AI, are trained to respond to user questions by scraping the internet for relevant information and assembling coherent answers, churning out convincing student essays, authoritative legal documents and believable news stories. But, because publicly available data contains misinformation and disinformation, some machine-generated texts might not be accurate or true. That has triggered a scramble to develop tools to identify whether text has been drafted by human or machine. Science is also struggling to adjust to this new era, with live discussions over whether chatbots should be allowed to write scientific papers or even generate new hypotheses.

Machine learning practitioners often end up tunneling on low-level technical details like model architectures and performance metrics. Could early model development instead focus on high-level questions of which factors a model ought to pay attention to? Inspired by the practice of sketching in design, which distills ideas to their minimal representation, we introduce model sketching: a technical framework for iteratively and rapidly authoring functional approximations of a machine learning model's decision-making logic. Model sketching refocuses practitioner attention on composing high-level, human-understandable concepts that the model is expected to reason over (e.g., profanity, racism, or sarcasm in a content moderation task) using zero-shot concept instantiation. In an evaluation with 17 ML practitioners, model sketching reframed thinking from implementation to higher-level exploration, prompted iteration on a broader range of model designs, and helped identify gaps in the problem formulation$\unicode{x2014}$all in a fraction of the time ordinarily required to build a model.