Virtual humans - H+ Weekly - Issue #409
This week - Italy bans ChatGPT; troubles in Neuralink; OpenAI invests into humanoid robot startup; mice with mini deer antlers; why should superintelligent AI keep humanity; and more!
At a first glance, Lil Miquela looks like any other Instagram influencer. A young girl sharing with 2.8 million followers her life and travels while hanging out with friends and having a good time.
But the thing is, Lil Miquela does not exist. She is a virtual person. And she is just one of many virtual influencers that popped up in recent years. Instagram alone has at least 35 verified virtual influencers, each with thousands (some even with millions) of followers.
The world of streaming has been a fertile ground for vtubers and vstreamers. We have seen creators like CodeMiko, ironmouse, Mori Calliope, Gawr Gura and many more gathering millions of followers and rising to internet fame and stardom. In all those cases, these virtual avatars are controlled and voiced by a real human. These humans are more or less playing their virtual personas.
However, recently someone replaced the human with a large language model and created neuro-sama - the first AI vstreamer.
Virtual humans are also present in the music industry. Last year, an AI rapper FN Meka made news when Capitol Records signed a contract with him. A month later, Capital Records dropped the AI artists due to FN Meka being built on top of racists and reductive stereotypes.
But if you want to see successful virtual artists in music and have a glimpse how they can be used in the future, I’d suggest looking at K-pop.
In 2018, Riot Games released K/DA - a virtual K-pop girl group with members being characters from League of Legends, voiced by real artists. Their first song - POP/STARS - was a massive hit and an internet sensation. It gathered 548 million views on YouTube and almost 300 million listens on Spotify.
K-pop’s next venture into virtual groups came in 2020 with the debut of Aespa - one of the most successful active K-pop groups. Aespa is a group in which each artist has their own virtual avatar, who are helping in promotions and advertising. However, from what I see, this concept seems to be abandoned.
And just a month ago, the first fully virtual K-pop group - MAVE: - debuted with _PANDORA which got over 20 million views on YouTube. Not a bad number for a debut song. MAVE: is using AI voice generators to interact with fans but the voices you hear in the song come from human performers.
Virtual influencers have become incredibly popular, getting millions of followers and views. Brands such as Coach, Dior, Balenciaga, Prada, Samsung and Calvin Klein have already worked with a virtual influencer.
A virtual influencer has several advantages over humans. They are available all the time and can be everywhere. They won’t make a PR problem with an inappropriate tweet. They won’t be tired or sick. The brands have full control over them. And according to HypeAuditor, virtual influencers have a higher engagement rate than human influencers.
The question is how long they can be relevant when the novelty wears off. Some people raise ethical questions about virtual humans and if it is acceptable for a non-human to advertise products and lifestyles to humans.
With their massive popularity, virtual influencers are not going away any time soon. And with the advances in deepfakes and generative AIs, soon they will be indistinguishable from real human beings.
🦾 More than a human
Neuralink’s FDA Troubles Are Just the Beginning
At a press event last November, Elon Musk promised that the first human trial of Neuralink’s brain implants would begin by the end of May this year. However, the recent news from the company makes that timeline doubtful. Neuralink failed to secure a green light from FDA due to “dozens of deficiencies”. On top of that, two other US government agencies are investigating Neuralink for alleged animal abuse and mismanagement and interstate portage of biohazardous materials.
The professor trying to protect our private thoughts from technology
In a new book, The Battle for Your Brain, Duke University bioscience professor Nita Farahany argues that such intrusions into the human mind by technology are so close that a public discussion is long overdue and lawmakers should immediately establish brain protections as it would for any other area of personal liberty. “I wrote this book with neurotechnology at the forefront as a wake-up call, but not just neurotechnology but all the ways out brains can be hacked and tracked”, says Farahany in this conversation with The Guardian. “Cognitive liberty is part of a much broader conversation that I believe is incredibly urgent given everything that is already happening, and the increasingly precision with which it’s going to happen, within neurotechnology”.
PiEEG Offers Affordable Brain-Computer Interface
If you ever wanted to build your own brain-computer interface, then check out PiEEG. It is an affordable, open-source board that can be connected to a Raspberry Pi 3 or 4, enabling RaspberryPi to receive 4 or 8 channels for connecting wet or dry electrodes that can measure biosignals such as those used in electroencephalography (EEG), electromyography (EMG), and electrocardiography (ECG). The project is currently raising funds on CrowdSupply and aims to cost $250 for a 4-channel board and $350 for an 8-channel board.
🧠 Artificial Intelligence
An Appeal to AI Superintelligence: Reasons to Preserve Humanity
If you were to convince a superintelligent AI to spare humanity, what arguments would you use? This is the question the authors of this article try to answer. Arguments they use vary from how would eliminating humans be received by other superintelligences to evoking the simulation hypothesis argument to how would that impact potential relationships with other intelligent agents (aliens included).
Introducing BloombergGPT, Bloomberg’s 50-billion parameter large language model, purpose-built from scratch for finance
Bloomberg introduces BloombergGPT, a large language model AI built specifically for the financial industry. Bloomberg claims this 50-billion parameter model outperform similarly sized open models. Other than a link to request a demo, the company does not disclose when the model will be available or what the price is going to be.
Italian privacy regulator bans ChatGPT
Last week, the Italian privacy regulator ordered a ban on ChatGPT over alleged privacy violations. ChatGPT will be banned until OpenAI respects GDPR rules. The authority said the company lacks a legal basis justifying "the mass collection and storage of personal data ... to 'train' the algorithms" of ChatGPT. The company also processes data inaccurately, it added. Reuters also reports that France, Ireland and Germany are also looking into ChatGPT over privacy and data security concerns.
Clearview AI used nearly 1m times by US police, it tells the BBC
Hoan Ton-That, the CEO of the facial recognition company Clearview AI, admitted in an interview with BBC that his company run nearly a million searches for US police. He also revealed Clearview now has 30 billion images scraped from platforms such as Facebook, taken without users' permission.
Beijing Welcomes Its New Robot Coworkers: China's Aging Crisis and Automation
Last year, for the first time since 1961, more people died than were born in China. With falling fertility rates and a quickly ageing population, China will have to find a solution to its incoming demographic crisis. One of the possible solutions is embracing automation and robotics as much as possible.
1X Raises $23.5M in Series A2 Funding led by OpenAI
OpenAI has quietly led a $23.5M investment round into 1X Technologies - a robotics company from Norway building humanoid robots. Their newest robot, Neo Eve, is scheduled to be released this summer.
Drone-on-Drone Combat in Ukraine Marks a New Era of Aerial Warfare
The ongoing war in Ukraine is the first armed conflict where both sides heavily utilise drones, from reconnaissance to combat missions. The proliferation of drones causes both sides to find new ways to deal with this threat. This resulted in the first instances of drone-on-drone combat where instead of bullets and missiles, drones use nets or ram into each other.
Mice grow mini deer antlers after stem cell transplant
Scientists in China have identified a new type of stem cell, which enables deer to regenerate their antlers year after year. When transplanted into mice, the rodents grew mini antlers, too, suggesting that the cells could trigger regeneration in other mammals — potentially even humans.
Bacterial ‘Nanosyringe’ Could Deliver Gene Therapy to Human Cells
Researchers found a way to utilise a bacterial nanosyringe to inject large proteins into human cells. The work could provide a way to deliver various therapeutic proteins into any type of cell, including proteins that can “edit” the cell’s DNA. “It’s a very interesting approach,” says Mark Kay, a gene therapy researcher at Stanford University who was not involved in the study. “Where I think it could be very useful is when you want to express proteins that can do genome editing” to correct or knock out a gene that is mutated in a genetic disorder, he says.
Your First Lab-Grown Burger Won’t Contain Much Beef
It turns out that producing a burger made entirely from lab-grown meat is too expensive to make it economically viable. Facing this problem, the companies working in the field of cultivated meat have started experimenting with mixing plant-based proteins with lab-grown meat in their products as a way to bring production costs down. The companies claim these mixed products taste the same if not better.
UK develops genetic early warning system for future pandemics
The Respiratory Virus and Microbiome Initiative is a project to create a system that would deploy DNA sequencing technology to identify all viral, bacterial and fungal species in a single sample collected from a nose swab from a patient. The system is to be used to pinpoint dangerous new variants as they emerge and act as an early warning system for new diseases and future pandemics. The team behind this initiative intends to make the technology cheap, easy to use and capable of being scaled up to provide global surveillance of a wide range of viruses.
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).
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.