This AI newsletter is all you need #28

What happened this week in AI by?Louis

Hottest News

1. AI-assisted code can be inherently insecure, study finds According to a new study involving several developers, code-generating algorithms were found to be more prone to being insecure compared to the “handmade” solutions created by the control group. Neil Perry, a Ph.D. candidate at Stanford and the lead co-author of the study, also advises developers to double-check generated code. Here’s why:
2. What we learned about AI and deep learning in 2022 As deep learning and generative models continue to make progress, the scientific community is divided on how to interpret these advances. To move forward, it is necessary to understand what is currently missing from AI systems and the architecture for the next generation of AI systems.
3. Machines that think like humans: Everything to know about AGI and AI Debate 3 The online AGI Debate #3, hosted by Vincent Boucher, the president of Montreal AI, and AI researcher Gary Marcus, discussed the lessons from cognitive science and neuroscience and the path toward achieving commonsense reasoning in AI. Here is a summary of the key points covered during the debate:

Three 5-minute reads/videos to keep you?learning

1. Here’s a no-code tutorial on how you can create 3D assets using Generative AI technology In this Twitter thread, @angrypenguinPNG shares a step-by-step tutorial on how to create 3D assets using generative AI technology. The tutorial covers all the prerequisites, essential tools, and the complete process for creating 3D models.
2. How to do video scripts on ChatGPT ChatGPT’s advanced conversational capabilities are making quite the buzz, but for it to produce accurate results, it needs precise and clear instructions. In this Twitter thread, @scottcmillard explains how to do it right and what factors to consider, including the inputs, variables, and more.
3. The Brief History of Artificial Intelligence: The World Has Changed Fast — What Might Be Next? In this article, Dr. Max Roser traces the brief history of AI systems’ language and image recognition capabilities and uses long-term trends to predict the future of AI. The article also discusses an important aspect of AI: creating a public resource to facilitate necessary public conversation.

Enjoy these papers and news summaries? Get a daily recap in your inbox!

The Learn AI Together Community section!

Meme of the?week!

Meme shared by dimkiriakos#2286

Featured Community post from the?Discord

DrDub#0108, an independent researcher, has just achieved over 2000 citations in the fields of natural language processing, AI, ML, and information retrieval. A big shoutout to Pablo! You can find a list of all the papers here and join in on the conversation here.

AI poll of the?week!

Join the discussion on Discord.

TAI Curated?section

Article of the?week

The Ever-evolving Pre-training Tasks for Language Models by Harshit Sharma

Self-Supervised Learning (SSL) is the backbone of transformer-based pre-trained language models. This paradigm involves solving pre-training tasks (PT) that help model the natural language. The author has made an effort to put all the popular pre-training tasks at a glance.

Our must-read articles

Lazypredict: Run All Sklearn Algorithms With a Line Of Code by Travis Tang

Markov Algorithm For Time Series by Ashutosh Malgaonkar

If you are interested in publishing with Towards AI, check our guidelines and sign up. We will publish your work to our network if it meets our editorial policies and standards.

Job offers

Senior Data Scientist @BluesInc (Remote)

Senior Machine Learning Scientist @Overjet (Boston, MA/Remote)

Senior Machine Learning Engineer @FarmWise (Paris, France/Hybrid)

Senior Machine Learning Engineer, Applied Computer Vision @Recursion (Toronto, Canada/Remote)

Cloud Machine Learning Engineer @HuggingFace (US/Remote)

Climate Modeling ML Internship @The Allen Institute for AI (Seattle/Remote Hybrid)