Discover more from Humanity Redefined
The Copyright Controversies of AI - H+ Weekly - Issue #412
This week - generative AI boosts workers' productivity; building a movement for longevity; "green" milk made from cells; people don't like Snapchat's AI chatbot; and more!
The AI revolution could not happen without three things - the development of brilliant algorithms, a rise in computation power and access to massive amounts of data to train those algorithms on. The last one - data - became again a centre of controversy last week.
The easiest way to get mountains of data to train an AI system is to go on the internet and just take everything you need without asking anyone for permission.
Not everyone is happy with this approach and some took action against it. Earlier this year, a group of artists sued Stability AI, Midjourney, and DeviantArt for the unauthorised use of copyrighted work to train image-generating AIs. They were soon after joined by Getty Image who sued Stability AI for infringing their intellectual property rights.
The recent explosion of large language models added its own set of copyright controversies to the mix.
Not long after that Reddit announced a plan to begin charging for use of its API. “Crawling Reddit, generating value and not returning any of that value to our users is something we have a problem with”, said Reddit co-founder and CEO Steve Huffman.
Media outlets under News/Media Alliance (which represent nearly 2000 publishers in the US) released their AI Principles, in which they clearly stated that “the unlicensed use of content created by our companies and journalists by GAI [generative artificial intelligence] systems is an intellectual property infringement: GAI systems are using proprietary content without permission”.
The question is where do we go from here? The lawsuits mentioned earlier haven’t been resolved yet so we don’t know what is allowed from the legal point of view.
What is most likely to happen is that other companies will take note of Reddit’s and StackOverflow’s examples and will try to protect their intellectual property by restricting access to it while also making some money out of it.
🦾 More than a human
The “Stepping Stone” Approach to Getting Longevity Drugs to Market
Proto.life asked some of the world’s leading experts in the field what it will take to get an FDA-approved pill that makes it possible to live a longer or healthier life. Though there was no consensus as to what the first longevity or “geroprotective” drug will be, several of these experts expressed confidence that it’s only a matter of time. Getting there won’t be easy, however.
▶️ Building a Movement for Longevity - Stephanie Dainow (1:31:29)
Here is a conversation with Stephanie Dainow - Executive Director at Lifespan.io - about longevity but not from the science but from a business point of view. Dainow shares her experiences from her efforts to promote longevity research, building the community and movement around it, and the ups and downs of being in the longevity business.
The first babies conceived with a sperm-injecting robot have been born
Right now, IVF labs are multimillion-dollar affairs staffed by trained embryologists who earn upwards of $125,000 a year to delicately handle sperm and eggs using ultra-thin hollow needles under a microscope. And there are startups which work to disrupt this with robotics and automation, potentially making the procedure less expensive and far more common than it is today. One such startup, Overture Life, even claims that two baby girls have been born after fertilization by a robot.
🧠 Artificial Intelligence
▶️ The Inside Story of ChatGPT’s Astonishing Potential | Greg Brockman (30:09)
In this TED Talk, Greg Brockman, President and Co-Founder of OpenAI, shows live on stage what ChatGPT (equipped with plugins) can do and shares the new paradigm of human-machine interaction where we say what we want and the machine figures out what needs to be done. In the second half of the video, Brockman is joined by Chris Anderson to speak about why OpenAI succeeded with a much smaller team than any tech giant, how new capabilities emerge from large language models, and the safety of advanced AIs and OpenAI’s approach to AI safety.
Snapchat sees spike in 1-star reviews as users pan the ‘My AI’ feature, calling for its removal
Snapchat joined the AI chatbot train and added My AI chatbot to the app. Snapchat users are not on board with this idea. The feedback is that the chatbot is invasive and raises concerns about the user’s private data.
Cortical Labs raises $10M for its Pong-playing stem cells which eventually could power AI
When every other AI hardware startup is looking at how to squeeze every drop of performance from silicon, Cortical Labs turned towards biology and builds AI chips powered by lab-cultivated neurons from human stem cells. The company has proven its idea works in 2022 by demonstrating that neurons in a petri dish can be encouraged to play the computer game Pong. With the latest investment into the company, Cortical Labs is seeking to use the chips in drug development but they are aiming beyond that - “if the number and complexity of these neurons are scaled, the end result would be familiar to us as fully embodied organisms such as a cat, dog, or human.”
▶️ Should We Fear Alien Artificial Intelligence? (13:56)
If you were an alien civilisation, what kind of artificial would you build? Would it run on pure mathematics? Would it be a standalone machine like our computers or will it be merged with aliens in some way? Would it be a general or narrow intelligence? John Michael Godier explores what an alien AI, programmed and trained without any human input, would look like.
First Real-World Study Showed Generative AI Boosted Worker Productivity by 14%
According to a new study by researchers at Stanford University and MIT, customer service workers at one Fortune 500 company became 14% more productive on average than those who were not, with the least-skilled workers reaping the most benefit. The most highly skilled workers saw little to no benefit from the introduction of AI into their work. The paper describing the study is available on arXiv.
▶️ Building Humanoid Robots (16:18)
This video does an excellent job of explaining why building a humanoid robot that can do tasks as well as a human is so hard - from solving the problem of bipedal motion to creating dexterous manipulators to equipping such a robot with an AI smart enough to understand what it is asked to do.
Quadrupeds Are Learning to Dribble, Catch, and Balance
IEEE Spectrum presents three interesting four-legged robots to be presented at this year's International Conference on Robotics and Automation (ICRA) - a very agile robot that can dribble, a robot that can catch balls into a net on its back and a robot that can walk a very narrow plank.
The ‘green’ milk made from cells
TurtleTree Labs is a small lab that works on creating a process to produce milk from animal cells in bioreactors. Their method of producing milk promises to be more sustainable, scalable and efficient than giant farms.
Organoid Intelligence: Computing on the Brain
We’ve made much progress in AI by taking inspiration from neurons. Some scientists then asked, why not use the actual neurons? The answer to this question led to the creation of organoid intelligence. Organoid intelligence is a collective effort to promote the use of brain organoids - tiny spherical masses of brain tissue grown from stem cells - for computation, drug research and as a model to study at a small scale how a complete brain may function.
▶️ The man who faked human cloning (1:04:44)
This video tells the story of a scientific scandal that shocked South Korea in the early 2000s. It is a story of how one person, Hwang Woo-Suk, rose from humble beginnings to become a leading figure in biotech and the hope of the entire nation. This is part one of two and covers the rise of Hwang Woo-Suk and ends when the truth of his breakthroughs in cloning is about to get exposed.
H+ Weekly is a free, weekly newsletter with the latest news and articles about AI, robotics, biotech and technologies that blur the line between humans and machines, delivered to your inbox every Friday.
Subscribe to H+ Weekly to support the newsletter under a Guardian/Wikipedia-style tipping model (everyone gets the same content but those who can pay for a subscription will support access for all).
This issue is also supported by my Patreons: whmr, Florian, dux, Eric and Andrew. Thank you for your support!
Thank you for reading and see you next Friday!
H+ Weekly is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.