Climate change data, Unreliable AI, Bard extensions, new AI courses and more...
I’ve compiled a short list of things that I have been reading and writing over the last few days and weeks.
New Data Shows that 2023 was the Hottest Summer Ever
(Medium - no paywall)
We develop visualizations in Python and Plotly to analyse the hottest-ever recorded temperatures over the period of June through August 2023. Full code is given in Python and Plotly with a downloadable Jupyter Notebook.
Why Do People Trust Unreliable AI
(Medium - no paywall)
Research suggests that Generative AI is extremely popular and relied upon even though, sometimes, it just makes things up. An article on Medium (no paywall).
Bard can now connect to your Google apps and services
Google’s LLM can now connect to your Google services like Gmail and Google Docs through the Bard Extensions. It can connect of several Google services like Flights and Hotels and can read your Gmail and connect to Google Drive so that you can access your data.
Advanced neural network techniques
The Elements of AI online course has been updated with a section on LLMs as part of its “Advanced Neural Network Techniques” section.
For those unfamiliar with the course, it is run by MinnaLearn and the University of Helsinki and is a free, self-paced online, course that explains the nuts and bolts of AI and includes practical exercises to test your understanding, many of them in Python.
AI Regulation: step with care, and great tact
Modular, the creators of the Mojo programming language have shared an interesting article about AI regulation, something that both the general public and industry need to think seriously about. The article puts forward the author’s view that we are perhaps worrying too much about the future impact of AI because the prospect of Artificial General Intelligence (AGI) is not imminent.