Last Week in AI #132: High-risk AI applications, AI can't be granted patents

Ongoing debates about how to classifying high-risk AI applications, U.S. federal court rules that AI can't be treated as inventors under current law, and more!

Sep 6, 2021

The Fight to Define When AI Is ‘High Risk’

In April, the European Union proposed the Artificial Intelligence Act, a large initiative to regulate how and which AI products can be used within the EU. One of its high profile proposal is to prohibit ‘high risk’ applications of AI such as law enforcement use of AI for facial recognition. More than 300 organizations have weighed on what ‘high risk’ should mean prior to the end of the comment period in August 8. For example:

The European Evangelical Alliance believes all forms of AI with the potential to harm people should be evaluated, and AI with the power to harm the environment should be labeled high risk, as should AI for transhumanism, the alteration of people with tech like computers or machinery.

More broadly, companies such as Facebook and Google have argued for less regulation, and civil society associations and humans rights associations such as Access Now and The Civil Liberties Union for Europe have fought for more stringent regulation. It will take years for this act to be passed and in effect, but much like the U’s GDPR will be a first-of-its-ind legislation with far reaching effects.k

Only Humans, Not AI Machines, Can Get a U.S. Patent, Judge Rules

A U.S. federal judge ruled in the case Thaler v. Hirshfeld that "a computer using artificial intelligence can’t be listed as an inventor on patents because only a human can be an inventor under U.S. law." This is the clearest position the government has taken yet on whether or not AI can be granted patents and be treated as an inventor under law. Opponents of this ruling cite that courts in South Africa and Australia have ruled in favor of AI as inventors, and that "this decision would prohibit protection for AI-generated inventions." It's still possible for AI to be treated as inventors in the U.S. in the future, if Congress makes new laws.

Other News

Research

DeepMind’s Collect & Infer: A Fresh Look at Data-Efficient Reinforcement Learning - "In a new paper, a DeepMind research team proposes a clear conceptual separation of the RL process into data-collection and inference of knowledge to improve RL data efficiency. "

GPT-3 mimics human love for ‘offensive’ Reddit comments, study finds - "Chatbots are getting better at mimicking human speech — for better and for worse."

Stochastic Parrots 🦜 and Shaky Foundations - "Meg Mitchell Speaks at Stanford HAI"

NVIDIA’s Isaac Gym: End-to-End GPU Accelerated Physics Simulation Expedites Robot Learning by 2-3 Orders of Magnitude - "A Nvidia research team presents Isaac Gym — a high-performance robotics simulation platform that runs an end-to-end GPU accelerated training pipeline. "

Facebook AI Open-Sources CO3D (Common Objects in 3D) Data Set For 3D Reconstruction in Computer Vision Research - "A new data set of nearly 19,000 videos capturing objects along with highly accurate 3D reconstructions."

AI Researchers Introduce A Graph Neural Network Estimator for ETA Prediction at Google Maps - "Web mapping services like Google Maps are excellent tools for dynamically navigating portions of the Earth, especially in urban areas."

Applications

Always there': the AI chatbot comforting China's lonely millions - "After a painful break-up from a cheating ex, Beijing-based human resources manager Melissa was introduced to someone new by a friend late last year."

Army Wants to Install Facial Recognition, Video Analytics at Child Development Centers - "The Army wants to use facial recognition and advanced machine learning algorithms to monitor kids at base Children Development Centers and plans to launch a pilot program at Fort Jackson in the near future. Army contracting officers posted a solicitation to SAM."

Jake Elwes: Constructing and Deconstructing Gender with AI-Generated Art - "Elwes’ latest venture, The Zizi Project, is an ongoing series of works applying diverse drag and gender fluid identities as training sets for positive AI outcomes."

Google’s New AI Photo Upscaling Tech is Jaw-Dropping - "Photo enhancing in movies and TV shows is often ridiculed for being unbelievable, but research in real photo enhancing is actually creeping more and more into the realm of science fiction. Just take a look at Google’s latest AI photo upscaling tech."

AI Gives Street Fighter Characters Human Faces - "While Street Fighter V’s character roster is (mostly) supposed to look human, everyone obviously has a cartoonish slant since the game has never tried to be photorealistic. Let’s imagine just for a moment, though, that Capcom was trying to make everyone look real."

NVIDIA’s latest tech makes AI voices more expressive and realistic - "The voices on Amazon’s Alexa, Google Assistant and other AI assistants are far ahead of old-school GPS devices, but they still lack the rhythms, intonation and other qualities that make speech sound, well, human."

AI startups claim to detect depression from speech, but the jury’s out on their accuracy - "According to founder and CEO Mainul I Mondal, Ellipsis’ technology is “science-based” and validated by peer-reviewed research. But experts are skeptical that the company’s product, and others like it, work as well as advertised."

Business

SenseTime's HK IPO could be a boon for China's controversial 'A.I. dragons' - "SenseTime Group, China’s most valuable artificial intelligence startup, has filed for an initial public offering in Hong Kong, potentially clearing the way for similar offerings by other Chinese A.I. ventures whose previous efforts to go public have been hindered by U.S. allegations of their ties to state security agencies."

Carbon Robotics raises $27M for its AI-powered weed destroying machine used by farmers - "Investors are putting more cash behind the self-driving weed-zapping machines made by Carbon Robotics that are attracting attention from farmers across the world. The Seattle-based startup just landed $27 million to help meet demand."

Rent-a-robot: Silicon Valley’s new answer to the labor shortage in smaller U.S. factories - "Silicon Valley has a new pitch to persuade small companies to automate: rent-a-robot."

These cute electric robots may soon deliver your dinner - "Kiwibot is a last-mile delivery service that has completed over 150,000 food deliveries using electric semi-autonomous robots. Yesterday, the company announced its official expansion into San Jose, Miami-Dade County, Pittsburgh, and Detroit."

Concerns

Tesla is ordered to turn over Autopilot data to a federal safety agency. - "The main federal auto safety agency has ordered Tesla to hand over a trove of data on its Autopilot driver-assistance system as part of an investigation into Tesla cars crashing into fire trucks or other emergency vehicles parked on roads and highways."

Amid Tesla’s Autopilot Probe, Nearly Half the Public Thinks Autonomous Vehicles Are Less Safe Than Normal Cars - "As the federal government investigates Tesla Inc. for crashes involving its vehicles using the “Autopilot” feature, a new poll indicates much of the public has safety concerns about autonomous vehicles, though Americans’ interest in the cars has risen slightly compared to a few years ago."

Even experts are too quick to rely on AI explanations, study finds - "The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Register now!"

GitHub Copilot Generated Insecure Code In 40% Of Circumstances During Experiment - "Researchers published a scholarly paper examining the security implications of GitHub Copilot, an excellent AI system now being used for code completion in Visual Studio Code. In various scenarios, some 40 percent of tested projects were found to include security vulnerabilities."

The time a human-driven car ran over an autonomous robot - "One of Starship Technologies’ delivery bots didn’t make it across the street"

Facebook Apologizes After A.I. Puts ‘Primates’ Label on Video of Black Men - "Facebook called it “an unacceptable error.” The company has struggled with other issues related to race."

Bias persists in face detection systems from Amazon, Microsoft, and Google - "A study by researchers at the University of Maryland finds that face detection services from Amazon, Microsoft, and Google remain flawed in significant, easily detectable ways."

Call center agents interested in, but wary of, automation - "Forty-one percent of call center agents say that they want AI to tackle complex calls. Yet 40% fear that AI will one day take away their jobs."

Analysis

How OpenAI Sold its Soul for $1 Billion - "The best intentions can be corrupted when money gets in the way. OpenAI was founded in 2015 as a non-profit company whose primary concern was to ensure that artificial general intelligence (AGI) would be created safely and would benefit all humanity evenly."

The Mouse Brain As A Gatekeeper For AI And Autonomous Cars - "Mice might be a crucial pathway toward figuring out the human brain. Mice might also be a crucial pathway toward devising Artificial Intelligence (AI) in machines. "

I Asked GPT-3 About Covid-19. Its Responses Shocked Me. - "Generative AI systems could guide future pandemic decision-makers"

Meet the women making waves in AI ethics, research, and entrepreneurship - "Women in the AI field are making research breakthroughs, launching exciting companies, spearheading vital ethical discussions, and inspiring the next generation of AI professionals."

Explainers

Bad Labels - "It turns out that bad labels are a huge problem in many popular benchmark datasets. "

Podcast

Check out our weekly podcast covering these stories!
Website | RSS | iTunes | Spotify | YouTube

Last Week in AI