Credit: Tatiana Shepeleva/Shutterstock.com.

Verdict lists five of the top tweets on artificial intelligence (AI) in Q2 2022 based on data from GlobalData’s Technology Influencer Platform.

The top tweets are based on total engagements (likes and retweets) received on tweets from more than 636 AI experts tracked by GlobalData’s Technology Influencer platform during the second quarter (Q2) of 2022.

Go deeper with GlobalData

Premium Insights

The gold standard of business intelligence.

Find out more

Related Company Profiles

View all

1. Pedro Domingos’s tweet on the large carbon footprint of developing AI models

Pedro Domingos, a professor of computer science at the University of Washington (UW), tweeted on machine learning (ML) consuming about 75% of the computing power of Google, a subsidiary of the technology conglomerate holding company Alphabet. Replying to this tweet, Alejandro Tejada C shared an article on training AI models generating as much carbon emissions as it took to manufacture and drive five cars over one’s lifetime. AI research has accounted for skyrocketing energy as well as financial costs, the article further detailed.

The reasons for ML and AI models using excessive energy were many. For instance, traditional data centre computation included social media, email, and video streaming. However, AI models consume more energy as they are computationally more intensive as they require to analyse tons of data and then learn from it, the article noted. As a result, training AI models is more inefficient when compared to human learning.

For example, modern AI uses artificial neural networks, which are mathematical computations that mimic neurons in the human brain. Every neuron connects to its neighbour via a network called weight. This network begins to adjust them with random weights until the output matches the correct answer in order to learn how to comprehend language.   

Username: Pedro Domingos

Twitter handle: @pmddomingos

Likes: 3,050

Retweets: 392

2. Andrew Ng’s tweet on how algorithms are evolving as AI advances

Andrew Ng, CEO of the software development company DeepLearning.AI, shared an article on understanding the origin of algorithms and how they are evolving as AI advances into society. According to him, many algorithms were invented long before digital computers. For example, linear regression guided navigators to follow the stars, and later biologists to find genetic traits in animals and plants, the article detailed.

Adrien-Marie Legendre, a French mathematician published the process of fitting a line to a set of points while trying to forecast the location of a comet, in 1805. Celestial navigation was then the most valuable science in global commerce at the time, similar to the AI of today. Logistic regression was later devised to understand how much of a given dangerous substance would be fatal. Logistic regression fitted the logistic function to a dataset to forecast the possibility, given an event led to a particular outcome.

ML, therefore, like many technical fields, evolved as researchers built on each other’s works. Some contributions remained and became the basis of future developments. Subsequently, everything ranging from a housing-price predictor to a text-to-image generator is built on algorithms, the article noted.

Username: Andrew Ng

Twitter handle: @AndrewYNg

Likes: 1,837

Retweets: 451

3. Ronald van Loon’s tweet on Netherlands AI-powered green village

Ronald van Loon, CEO of the Intelligent World, an influencer network that connects businesses and experts to audiences, shared a video on Netherlands building an AI-powered green village. Real estate firm ReGen Villages will be cultivating organic vegetables on vertical farms, and will recycle the waste into animal feed and fertiliser, the video demonstrated. The village will also generate its own solar and biogas power, while AI will manage its infrastructure. The company plans to design villages for every climate, especially for growing populations like in rural India and sub-Saharan Africa, the video showed.

About ten billion people on the Earth would require a home by 2050. The company believes that de-urbanisation is part of the solution as cities get overcrowded and expensive, and as people try to reduce their carbon footprint. For now, about 203 homes are expected to be built across 50 acres in the Netherlands.

Username: Ronald van Loon

Twitter handle: @Ronald_vanLoon

Likes: 93

Retweets: 66

4. Pieter Abbeel’s tweet on Skydio’s advanced AI-driven drones

Pieter Abbeel, a professor at the University of California, Berkeley, shared a podcast on the  US drone manufacturer Skydio developing advanced AI for drones. Adam Bry, co-founder and CEO of the unicorn start-up stated how the company is producing autonomous drones. The drones look like a typical drone with four motors, with a camera in front, but what makes them unique is their autonomy system, Bry added.  

The drones, as a result, have three navigation cameras on the top and three cameras at the bottom, making it a total of six navigation cameras with about 45 megapixels of navigation data, Bry stated. They also have an NVIDIA TX2 computer embedded inside that is running all of the AI and autonomy algorithms, and all of the hardware is designed to enable the autonomy software, which is where the company has made its most investments, the discussion detailed.

Username: Pieter Abbeel

Twitter handle: @pabbeel

Likes: 78

Retweets: 20

5. Catherine Adenle’s tweet on AI use cases in everyday life

Catherine Adenle, an AI and ML specialist, shared an infographic on AI applications in everyday life. For example, smartphones’ face ID is powered by AI. According to a research, 46% of Americans looked at their phones first thing in the morning. Likewise, AI influenced the contents viewers see on their social media that is determined by users’ tastes, preferences, an online browsing history, the infographic noted.

Smart homes are also powered by AI, while AI in navigation apps such as the Google Maps delivers data of the traffic flow around routes to be taken by users, suggesting even faster routes to reach a destination. Digital assistants are another use case of AI, where apps such as Siri, Alexa, and Google Assistant, can perform tasks as per the users voice command. The infographic further detailed how AI is used to filer spam emails, in text editors, and e-payment systems used by banks.

Username: Catherine Adenle

Twitter handle: @CatherineAdenle

Likes: 74

Retweets: 62