Let's Talk: The Dawn of AI Policy and the Decline of Meta

In this month’s installment of our Let’s Talk podcast series, AMT Lab staff Maraika Lumholdt and Natalie Larsen continue discussing the ever-evolving world of technology and privacy , especially in the realms of theater and music. The AI Bill of Rights, an important new blueprint for the future of technology policy, is also examined.

Listed in order of discussion are the articles referenced:

Biden’s AI Bill of Rights Is Toothless Against Big Tech

SEC Probing Bored Ape Creator Yuga Labs Over Unregistered Offerings: Report

Meta’s VR Headset Harvests Personal Data Right Off Your Face

Is Tech Changing the Legacy of Theatremakers?

Record Labels Say AI Music Generators Threaten Music Industry:



INTRO: Every day, every every podcast you do, every new tech news story comes out-- I feel like we're closer closer to living in a real life Black Mirror episode. Yeah. Welcome to another let's talk episode of Tech in the arts, the podcast series on the arts management and technology laboratory. The goal of our Let's Talk series is to exchange ideas, bring awareness and stay on top of the trends. My name is Maraika Lumholdt, host of tech in the arts and today I'm joined by our AMT Lab staff writer and lead researcher Natalie Larsen to talk about current news that has been happening in the month of October 2022. The topics we'll cover today are issues of data privacy in AI technology, the ongoing legal battlefield of NFT's and digital strategies in theater.


 

Natalie Larsen 

So my first story here is on a new blueprint for an AI Bill of Rights, the White House Office of Science and Technology Policy, or OSTP. They just released this on October 4. This bill contains five key principles. And those are that people have the right to control how their data is used, opt out of automated decision-making live free from ineffective or unsafe algorithms when AI is making a decision about them, and to not be discriminated against by unfair algorithms. I feel like this bill is incredibly important given all the recent controversy surrounding AI. It doesn't actually have the force of law behind it. And as the name suggests, it is just a blueprint, so it doesn't have any legal enforceability. One of the biggest criticisms that I found against this bill, however, is that it's only aimed at the federal government, and not really at the big tech companies who really have all the power when it comes to AI. It's also been said that the tenants of this bill are too vague and can't really be enforced or make a difference in people's lives. But the OSTP has intended to put this bill into practice and said that this is really just the first step in kind of regulating AI. They built this bill, or they built this blueprint from talking to individuals in the general public, as well as large companies like Microsoft, AI auditing startups and human rights groups. So they're really seeking a lot of input from different groups, different populations to make sure that this bill is equitable, and that it can make a real impact. However, it's also interesting to note that compared to the EU, our government is pretty far behind when it comes to regulating AI. The EU has been debating the Artificial Intelligence Act. And it's been quoted as being one of the first major policy initiatives worldwide, focused on protecting people from harmful AI. So while this bill is in its final stages of drafting, there is an expected decision to come mid November. And we'll see if this impacts efforts in the US similar to how the GDPR spurred on privacy regulation here.

 

Maraika Lumhodlt 

What else is new that the EU is like leagues ahead of us in terms of privacy? We see this all the time. I'm super curious to see what that decision will be. And I'm curious, too, in your research, did you see any feedback on the US' version critiques from big tech at all? I know it sounds like they were consulted and some of the language in the proposed bill. But has there been pushback on their end that it would harm what they're doing?

 

Natalie Larsen 

Not really that I've seen, I think when it comes to Microsoft, specifically, they're really in favor of imposing some sort of regulation to protect the general public. But as other companies, I haven't really seen anything. So yeah, it sounds like for the most part, all these different groups are in favor of this. I'm certainly in favor of it. And I really hope to see it kind of pushed forward.

 

Maraika Lumhodlt  

Yeah, definitely. It's always interesting to see how or when government which I would say is notoriously slow, matches up against a notoriously fast moving side of technology. Interesting to see how they'll keep up. I know, on the EU side, it's targeting big tech, less so than we are in the US and it's less focused on government. So two very different approaches to regulating AI. And I think as time passes, and these bills get further along, we'll, we'll see how it shapes up in the global sphere of AI regulation.

 

Natalie Larsen 

Right, and maybe because Microsoft was consulted, that they kind of played a part in it not being targeted, so to speak. But hopefully, again, like they said, This is really just kind of the first step so hopefully something more will come out of this in the future.

 

Maraika Lumhodlt 

Yeah. Very interesting. And speaking of government intervention, I think a company that in the NFT sphere is very well known Bored Ape, I think one of the more prominent NFT companies, has recently been probed by the US Securities and Exchange Commission. And our last let's talk episode, we talked about some new state laws that were coming into effect that would potentially be taxing NF T's like other assets. And so that, of course, is still new, and there's efforts there to regulate. But in this specific case, the SEC is investigating Bored Ape, well actually Bored Ape's creator Yuga Labs for potentially violating federal law for sales of its digital assets. So the specific probe is examining whether NFT's are similar enough to stocks, which would then make them subject to the same disclosure rules that stocks currently are. So Yuga Labs has stated that it's "fully cooperating with the SEC in hopes of partnering with the rest of the industry and regulators to define and shape the burgeoning ecosystem." That's a direct quote from them. But overall, I think this is just an example of the latest attempt by the SEC to regulate the crypto market, I think, notoriously unregulated, or I think the NFT industry has notoriously been unregulated, and has faced a lot of backlash because of this. So this is one way that SEC can lean into I think its current jurisdiction and do what they can to attempt to regulate NFTs in a certain way. But as we talked about last month, there's more and more potential for specific laws written to adapt to NF T's which are, you know, still relatively new and don't have the same laws that we've been applying to other similar assets and more traditional assets like stocks, since those came to be? So I thought this was interesting.I feel  like I've said that 10,000 times.

 

Natalie Larsen 

Yeah, no, it's definitely an interesting story, Marika, and I'm not personally as familiar with regulating securities and other financial information. So for the listeners out there, if you have any thoughts or opinions about regulating the world of web3, and you're more familiar with this sort of terminology, and you want to share your thoughts with us, please comment on this episode or email us get in touch with us through social media. Yeah, we'd be curious to see what you think. So taking us into our next story, we have another controversial story involving meta, which is a huge shocker they've come under a lot of public scrutiny in the past few years. So back in 2021 Meta, and I believe they were known as Facebook at the time announced it would be deleting face recognition data extracted from images from over 1 billion people, and that it would stop offering to automatically tag people in photos and videos. And they received a lot of criticism for this. A lot of people basically just call this a PR stunt. And that critique kind of proved right--Their new VR headset QuestPro, which contains internal cameras that track user's eye movements and facial expressions, as well as exterior cameras that capture leg movement, which is kind of wild, has raised a lot of different privacy concerns. So Mark Zuckerberg has stated that these cameras allow the avatar in the metaverse to look more realistic and less so like a cartoon. To me, they all just look like versions of the Sims. It's just it's too comical for me, I can't get over it. But a lot of experts have commented on their concerns about this headset being used to track a person's emotional state and then using that information to exploit them. For example, information collected on a user would allow such targeted ads to users that they're not even aware that they're being advertised to which is a huge ethical issue to me of autonomy and free will. Meta has said that they will be deleting images collected from the interior cameras, after they have processed that information. That data is still broadcast to outside companies and is subject to us on their own terms and privacy policies. So our information if it's collected through this headset is not necessarily protected by Meta even.

And I think this is a really a really critical time for meta. They have not been doing well just given all of this controversy in the past year. And now with this VR headset happening. There was an article that came out recently that found that their shares have actually fallen 24% in value. They're at their lowest point in the past four years nearly and in an article That was just put out by Vice. They found that their share prices have actually fallen by 70% just this past year and they credit that to Zuckerberg's, quote "obsession with the metaverse." Several investors have already pulled out and it's just a really interesting story to me because it feels it really does feel like kind of the downfall of meta Facebook, Mark Zuckerberg. That whole kind of empire.

 

Maraika Lumhodlt 

Wow. I never thought we would see that day. You know, I feel like they're betting everything on their Metaverse and that's everywhere they're headed and yeah, it doesn't seem to be in line with what I was gonna say Americans what people worldwide are in line with and what seems right to them. I have admittedly I have a Quest headset and now I am debating burning it so they don't collect all my data. Every day, every every podcast you do every new tech news story comes out. I feel like we're closer and closer to living in a real life Black Mirror episode.

 

Natalie Larsen 

Yeah, very fun. So fun.

 

Maraika Lumhodlt 

Well, on a somewhat happier note, but not without controversy. My next story is about how technology is changing theater. So theater, something near and dear to my heart. This is specific to Australia where we're looking at a specific trend, there's been a push towards digital documentation of theater, that the popularity of this is rising significantly in Australia. So what digital documentation would do would allow for greater sharing amongst practitioners have what's usually an ephemeral art form, right? Like you come to a theater, you watch, play or musical, whatever it is, and then you leave it's temporary. But it also allows for new points of engagement with audiences that wouldn't be possible in a traditional theater setting. There's an Australian Theatre collective called Pony Cam. And they experiment with ways in which their audiences and emerging independent theatre makers can engage with TheatreWorks and put them on their head, flip them around, see what new forms are most effective. And Pony Cam says that digital documentation helps emerging theatre makers learn from those who've been practicing it for a long time. And it also makes the ability to engage with both the major players in the theater space and independent artists. So it improves accessibility there and helps create the connections between those institutions of very different sizes and artists as well.

So creators in this push are being encouraged to document their work for archival purposes, for grant writing, and accessibility. And digital documentation can also be used to inspire artists, which I think is an important point who can make valuable contributions to the field. So if these independent artists are seeing more art in different forms, and getting access to it, that can serve as a launching point for their careers, and so they can help contribute to the overall theatre landscape as well. This is not without controversy, as I think technology, usually garners in the art space, so there are a lot of people who are expressing concerns about the balance between live theater, and recorded theater productions. I think something that I hear a lot, and I think I believe, to an extent is that the ephemeral quality of live theater is really important to the experience. And by recording a theater piece and removing that ephemeral quality, it's losing some of the value that the production could have otherwise. So really, the value of these recordings is only as the recordings knowing that they can never really capture the liveness of the experience.

So I do think that's a very important distinction. I personally agree with the purposes of archiving it, like there's an amazing resource in a library at Lincoln Center, where every Broadway show is recorded, and people can go watch. And, you know, that just creates a kind of posterity that's only possible with technology, in this case and with recordings. But I do think there is a pretty strict line between those kinds of internal occupation archival pursuits, and distribution for the form of getting more eyes on this project, getting a bigger audience that you can achieve within the boundaries of traditional theatre. And a big piece of that line or something that's majorly impacted is union agreements. So theater is a heavily unionized space. And the Actor's Equity Union, for example, has very strict rules on what can be recorded and the purposes that it's for. So, you know, a staged theatre production can have short snippets recorded to be used in publicity, and it comes with all sorts of things on top of that, like extra payments scheduling a bunch of different requirements, but it doesn't overtly allow a full theater piece to be recorded and then just distributed. Increase the audience that seeing this theater piece. It's it's allowed for that internal purpose, but not necessarily for external viewership. So, of course, COVID has created a lot of questions around this. And I think digitization within theater has gained a lot of use cases because of COVID. But no matter where you're at, whether it's with this rise in Australia, or here in the US, chances are there's a union agreement that you'll have to work through to put those practices into place. So I do think that's an important caveat. But I love finding out about new ways that digital media or filming in this case can help theater and I do understand why they're pushing for this in Australia.

 

Natalie Larsen 

Yeah, I'd love to sit down on one of those union negotiations and discussions just to hear kind of all the different points that are raised about this. I think all that stuff is really interesting.

 

Maraika Lumhodlt 

Yeah, I agree. And it is one of the most nuanced conversations I can think of, there's so many different points on both sides of the table. And I'm not intimately familiar with how the Australian actor’s unions work, but I imagine there are conversations around this happening there as generally, that is what has come to be.

 

Natalie Larsen 

Alright, so taking us into our final story, the Recording Industry Association of America or the RIAA, has released a statement, basically saying that any AI software that uses its member’s music is not authorized to do so. The RIAA submitted a review of what they deemed to be harmful sites where AI used a copy or some portion of a recording to generate a very similar track. So the story was originally posted by torrent freak who uses the term AI piracy to describe this kind of the same issue. Some of the websites that the RIAA mentioned were acapella extractor removed vocals, and song master. Basically, this AI software allows for songs to be kind of split up based on tracks or instrumentation. And then they're basically just remastered and remixed into a version that sounds very much like the original, but kind of enhanced in some way. But it's all still AI-generated. So of course, as with the visual art world, AI-generated work, relies on data sets of inputs that contain collections of original works by human artists.

The RIAA has even compared this issue in the recording industry, to the use of DALLE-2, which of course uses a dataset of visual art to generate images, this can be very harmful to musical artists as well as visual artists. I thought it was really interesting, just because last time we focused a lot on the visual art world in AI. And this just kind of goes to show that basically no artist is safe. That sounds pretty bleak. But no artist is really protected against AI, just with the current state of policy and regulation or AI, music. Software sites don't really have a huge following. The RIA A is taking this issue very, very seriously. They have a history of really going after copyright infringement or piracy cases. So their actions going forward could really determine just kind of the future of this kind of AI-generated musical works.

 

Maraika Lumhodlt 

That's interesting. I think I'd only really thought of these kinds of concerns around visual art, as you said, with DALLE Mini and those more, maybe prominent user-facing AI experiences that we have the most exposure to in our everyday lives, but makes total sense that, of course, the same concerns apply to recorded or music, art as music as well. And I think RIAA has really been a leader here. Remember earlier this year, they were heavily involved with taking down a website called hitpiece that had posted a bunch of NFT's which were NFT versions of songs that were posted and traded without any of the copyright owner's permission. So props to RIAA for being so on top of it. And hopefully in this case, as you said, these sites aren't using or aren't gaining a ton of users every day, but maybe this can help get ahead of any bigger concerns and set a precedent that the music industry will not tolerate any of this.

 

Natalie Larsen 

Yeah, I really hope this kind of sets a precedent for all art forms that human artists need protection, or they at least need compensation if their works are going to be used in training these AI technologies.

 

Maraika Lumhodlt 

Totally, I feel like an undercurrent of a lot of what we talked about around AI and other technologies and how they impact our decision. It needs to be a supplemental tool for artists and not a complete replacement. Or if it is, can we even think of that as art? Right, but ultimately, and ideally, we're in a situation where artists are treated fairly, they have access to these technologies if they want to use them to enhance their art, but they're still artists and credited as such at the end of the day.

 

Natalie Larsen 

Right. Yeah, I totally agree with you. I think there's nothing wrong with using this kind of stuff as a complement to your art. But you get into some really murky territory, especially when just non musicians or nonartists start kind of playing with this technology and putting works out there. And that kind of starts to overshadow the stuff that actual artists and musicians are doing.

 

Maraika Lumhodlt 

Definitely, it'll be interesting to see if this initial harsh action and not harsh, I should say justified, but swift action from RIAA will help prevent that. And I hope we don't see many more instances of this, but we'll see.

 

Natalie Larsen 

Thank you for listening to tech in the arts. Be on the lookout for new episodes coming to you very soon. If you found this episode, informative, educational or inspirational, be sure to send this to another arts or technology aficionado in your life. You can let us know what you think of this podcast by visiting our website amp lab.org That's AMT-lab.org or you can email us at amtlabcmu@gmail.com. You can follow us on Twitter and Instagram at Tech in the Arts, or Facebook and LinkedIn at Arts Management and Technology Lab. We'll see you for the next episode.