Tech Brew Ride Home - Wed. 04/12 - Twitter Inc. No Longer Exists?

Episode Date: April 12, 2023

Elon Musk did a big spontaneous interview with the BBC last night on Twitter Spaces, but here’s two things that didn’t make the interview: Twitter appears to be working on generative AI for reasons yet unknown and Twitter Inc. doesn’t exist as a company anymore. Global VC funding for crypto startups fell to $2.4 billion in the first quarter of 2023, down 80% from an all-time high in 2022. Video game illustrators in China say AI image generators are causing them to lose their jobs. And Reddit moderators saying AI-generated spam is rapidly mounting as they brace themselves for a spam apocalypse. Links: Elon Musk is moving forward with a new generative-AI project at Twitter after purchasing thousands of GPUs (Insider) Twitter Inc. has been merged with X Corp. and “no longer exists,” Elon Musk’s company says in a court filing. (Slate) Six things we learned from Elon Musk interview (BBC) Crypto VC Funding Plunges by 80% in Dire Quarter for Startups (Bloomberg) AI is already taking video game illustrators' jobs in China (Rest of World) Reddit Moderators Brace for a ChatGPT Spam Apocalypse (VICE) NPR to Suspend Twitter Use After ‘Government-Funded’ Label (NY Times) Learn more about your ad choices. Visit megaphone.fm/adchoices

On April 4th, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. Welcome to the tech meme ride home for Wednesday, April 12th, 2023. I'm Jackson Bird, filling in for Brian McCullough. Today, Elon Musk did a big spontaneous interview with the BBC last night on Twitter spaces, but here's two things that didn't make the interview. Twitter appears to be working on generative AI for reasons yet unknown, and Twitter, Inc. doesn't exist as a company anymore. Global VC funding for crypto startups fell to $2.4 billion in the first quarter of 2023, down 80% from an all-time high in 2022. Video game illustrators in China say AI image generators are causing them to lose their jobs.

And Reddit moderators are saying AI-generated spam is rapidly mounting as they brace themselves for a spam apocalypse. Here's what you miss today in the world of tech. Last night, Elon Musk did an impromptu interview with BBC reporter James Clayton on Twitter Spaces, and it was pretty frank, seemingly in part because, as Musk said, at one point, he considered it as much a chance for him to interview the BBC as them to interview him. A few of the big headline-making statements include his admission that at a certain point he only went through with the purchase of Twitter eventually because he knew a court was going to make him do it. And he insisted on the joke that he's been making for a while that his dog, Flokey, is Twitter's CEO, not him.

There was a tete-a-tete about misinformation that lots of people are talking about, but I want to pull away those headlines for some juicier Twitter news that Musk is not being as candid about. sources say that Twitter is working on a generative AI after having bought about 10,000 GPUs and poaching some engineers from deep mind. Quoting insider, Musk's AI project is at an early stage, according to the people familiar, the one person said the purchase of so much additional computational power showed he's committed to its development. Musk, however, has criticized the recent development of generative AI, saying the technology, is powerful and needs regulation to ensure it's operating within the public interest. Musk's AI project at Twitter involves a large language model, according to one of the people familiar.

Twitter has massive amounts of data that could train in LLM. For instance, ChatGPT's creator, OpenAI, previously had access to Twitter data for training purposes, though Musk said he put a stop to that in December. Musk has also brought new AI talent to Twitter by early March. March, he had officially hired the engineers Igor Barbushkin and Manuel Croix from Alphabet's AI research subsidiary Deep Mind. Since at least February, Musk has approached people in the AI field, including Babushkin, about starting his own AI endeavor to rival chat GPT, the information reported, end quote. Insider throws out some theories for what Twitter could be using the Gen.

AI4, maybe improving search, maybe improving advertising. Brian has a bit of a different theory that he sent over to me on Slack, so here's Brian's analysis. One of the longstanding theories about Elon and Twitter was that he would try to fix it and make it better. If that didn't work, he'd try to turn it into a subscription product hybrid. If that didn't work, he'd try to make it into an Asia-style super app with payments and everything. And when all of that didn't work, he'd try to make it into an Asia-style super app with payments and everything. And when all of that didn't work, he'd just shut it down and be like, hey man, I tried everything. That thing was so broke, even I couldn't fix it. But, okay, new theory. He spins up a chat GPT competitor in six months. The hope here is that he can use the hype around

that to get new equity investments or even try to spin it out as an IPO. Maybe he keeps the legacy social network running on autopilot, but can be like, sure, we have that, but here's the future. Quick, give me a $30 billion valuation. I mean, it's a long shot, but it's plausible. But it also makes the likelihood of him shutting down Twitter, declaring it too broke to fix and washing his hands of it even more possible. If he can spin up an LLM that gets buzz, why bother with Twitter at all anymore?

Oh, and then there's this, end quote. This being a court filing from last week on April 4th, revealing that Twitter Inc. no longer exists. It has merged into X-Corp. This court filing comes from a lawsuit from Laura Lumer against Twitter after she got banned from the platform in 2019 and accused the site of violating federal racketeering laws. In the most recent filing, X-Corp was listed as the defendant in the suit instead of Twitter, with an update regarding the merger.

Quoting Slate, Elon Musk, who owns Twitter as well as X Holdings Corp, has not yet revealed this merger to the public, but it appears to have been on his mind since he first plotted his purchase of the social media company. In April 2020, Musk registered X-holdings 1, 2, and 3 in Delaware, three separate companies designed to facilitate his purchase of Twitter. As the merger agreement stated, X-holdings 2 would cease to exist as a functioning entity after merging with Twitter. So the official structure after Musk's takeover was X-holdings-1 oversees Twitter, Inc, while X-holdings-3 handles the cash. The Nevada filing

show this exact arrangement is no longer in effect. According to the Nevada Secretary of State's online business portal, Elon Musk registered two new businesses in the state on March 9th, X Holdings Corp and XCorp. Then on March 15th, Musk applied to merge those Nevada businesses with two of his existing companies, X Holdings 1 with X Holdings Corp, and Twitter, Inc. with XCorp.

In the latter's case, the articles of the merger mandate that XCorp, fully acquire Twitter, meaning that for all intents and purposes, Twitter Inc. no longer exists as a Delaware-based company. Now it's part of X-Corp, whose parent company is the $2 million X Holdings Corp. After completing the Twitter takeover in October, Musk claimed this purchase would accelerate the creation of an everything app that would be named, you guessed it, after the letter X. Such a super app, Musk has stated, could resemble something like China's WeChat, the combination messaging, social networking payment app that boasts a billion users. Twitter's functions would play an important role in this mega-size app.

It's also possible that X is meant to be more of an everything company than an everything app. In late 2020, YouTuber and longtime Tesla investor Dave Lee tweeted that Musk should form a holding company called X to serve as the parent company of Tesla, SpaceX, NeurLink, and Boring company. Musk's response, good idea, end quote. Global VC funding for crypto startups fell to $2.4 billion in the first quarter of 2023, down 80% from an all-time high of 12.3 billion in Q1 of 2022, and is now at the lowest level since 2020. This, according to data from Pitchbook, quoting Bloomberg, the drop is not a surprise, said Pitchbook Cryptoanalyst, Robert Lay, who noted that venture investing has dwindled across the board this year.

In addition to rising interest rates, the first quarter also saw the unraveling of Silicon Valley Bank, an institution widely relied on by venture-backed companies. There's still a lot of fear about what's going to happen since the macro environment is still very uncertain, Lay said. Crypto has had its own special challenges. Lay said the collapse and bankruptcy of crypto exchange FTX has helped slow down the pace of funding rounds and has reinforced the need for due diligence. Rather than rushing into deals, VCs are conducting months of research and asking founders more questions before deciding whether

to back a startup. It's not going to be based on FOMO or what other investors are doing, Le said. Despite the size of the downturn, there was some good news for crypto startups in the pitchbook data. On a month-over-month basis, crypto venture investing actually increased in February and March, meaning that the worst of the funding drought may be over, lay said. Venture investors have money in their coffers and that there's still interest in backing crypto infrastructure startups, data analytics firms, and developer platforms. We'll slowly start seeing investors get more comfortable, lay said, end quote. The use of generative AI in video game production is soaring in China, and illustrators in particular

are saying it's causing them to lose their job. or fear for their futures. Quoting, rest of world, recent breakthroughs in AI image generation, with the release of programs such as Dolly 2, Mid Journey, and Stable Diffusion in 2022, have enabled users to produce impeccable drawings from text prompts. In the past few months, Chinese video game companies, from tech giants like Tencent to indie game developers, have begun using these programs to design and create video game characters, backdrops, and promotional materials. AI is developing at a speed way beyond our imagination.

Xu Ying Ying, illustrator at an independent game art studio in Chongqing, told rest of world. Xu's studio produces designs for major game developers in China. Five of the studio's 15 illustrators, who specialize in character design, were laid off this year. And Xu believes the adoption of AI image generators was partly to blame. Two people could potentially do the work that used to be done by 10, she said. A game artist in Guangdong speaking on condition of anonymity for fear of being identified by her employer said that previously employees could draw a scene or a character in a day. Now, with the help of AI, they could make 40 a day for their bosses to choose from. I wish I could just shoot down these programs, the artist told rest of world after getting off work late one night.

She said fear of impending layoffs had made her colleagues more competitive. Many stayed at work late, working longer hours to try to produce more. AI made us more productive, but also more exhausted, she said. And that is personally one of my bigger fears about AI and jobs. It's not going to lead to a culture shift of more leisure time. When has any tech development with that rosy promise followed through equitably? Der Obasanyo said on Twitter, quote, There are a number of jobs that AI will

will make obsolete, which will mirror the language translation industry. Tasks which used to take a skilled human days now take seconds. So you go from needing a full staff to just one or two people to fix up errors, end quote. The AI generators aren't perfect yet. One artist told rest of world that they struggle within more niche styles, and some companies like NetEase say they are only trained on the company's own, licensed or proprietary resources. Still, artists say they went from designing and illustrating complete pieces to being commissioned just to fix small AI errors, being paid a fraction of their previous rates, and having much of the creative artistry they're passionate about stripped away.

Quoting again, Jeffrey Ding, an assistant professor at George Washington University, who studies development of AI in China, said the advancement of AI could open up competition and create new opportunities, but it could also eliminate a way. wide range of white-collar jobs currently done with computers. The reality might be that AI will displace a lot of jobs, not just artists, but lawyers and writing services, ding-told rest of world. End quote. Well, it doesn't solve the problem of creative fulfillment, but here's another job us lowly humans may be cut out for, manually filtering out the spam that AI bots churn out online. at least until other bots figure out how to do that better than us, or rather if the bots to control the spam can keep up with the advancements in bots that create the spam and outwit the filters.

But for now, moderators on Reddit, at least, have been seen an increase in bot-created spam that they say, to take care of, will, quote, require a lot of human labor. End quote. Human labor at the moment being performed by moderators who are all volunteers. Quoting Vice, Shortly after ChatGPT launched, the 2 million-strong Ask Historians subreddit

was experiencing 5 to 10 chat GPT posts per day, says Sarah Gilbert, one of the subreddit's moderators and a postdoctoral associate at Cornell University, which soon ramped up as more people found out about the tool. The frequency has tapered off now, which the team believes may be a consequence of how rigorously they've dealt with the AI-produced content, Even if the posts aren't being deleted for being written by chat GPT, they tend to violate the subs standards for quality.

The moderators suspect some chat GPT posts are aimed at testing the mods or seeing what the user can get away with. Other comments are clearly part of astroturfing and spamming campaigns or engaged in karma farming, where accounts are set up to accumulate upvotes over time, giving them the appearance of being authentic so that they can be deployed for more nefarious purposes later on. Several hundred accounts have already been removed from the site, and more are being discovered daily, they said, adding that most of the removals are being done manually because Reddit's automated systems struggle with AI-created content.

Reddit declined to offer any comment on this. In February, Ask Historians and several other subreddits were hit by a coordinated bot attack using ChatGBT. The automated system of bots was caught inputting questions asked on Ask Historians into ChatGBTGT and then spitting out responses through an army of shill accounts, says Gilbert. Identifying that the bots spamming answers were produced with ChatGPT wasn't the problem. It was that they were coming in so fast and so quick, says Gilbert. At the height of the attack, the forum was banning 75 accounts per day over the course of three days, end quote.

As a moderator of our technology explains, by the time a bot is caught by Reddit's auto-antyspam systems, it's usually too late because the bot has already served its purpose. Its purpose often being advertising, often for adults-oriented drop-shipping products and scams, not so much more nefarious political manipulation yet. The bots are getting better, however, at resisting detection, especially the karma-farming ones. In anonymous R, Cybersecurity moderator told Vos, quote, our problem isn't necessarily what we've found so far, but what we've missed, end quote.

And it's less a huge problem now and more a concern for how much greater a problem it could become. Gilbert says Reddit themselves have a huge incentive to get a handle on this now to take some of the work off the volunteer moderators. Quote, they want people to read their ads, right? It's not like Google's AI chatbot Bard is going to buy anything, end quote. And the our cybersecurity moderator added to, to that, telling Vice that what GPT will really change is social media, saying, quote, I'd bet the farm that traditional social media has a finite lifespan, largely because inauthentic content is becoming so realistic and cheap to make that we're going to struggle

to find who's real and who's a bot, end quote. One more piece of Twitter news that broke while I was writing the show this morning, NPR has announced that it will suspend all Twitter use. for its organizational accounts. This comes, quote in the New York Times, a little over a week after the social network designated the broadcaster, U.S. state-affiliated media. Twitter has since changed the label on the NPR Twitter account

to government-funded media, a label it also gave to the BBC, the National Broadcaster of Britain, end quote. NPR is editorially independent and receives, quote, less than 1% of its annual operating budget in the form of grants from the government-funded corporation for public broadcasting and other federal agencies and departments, end quote. So, NPR has said, quoting a statement by Chief Communications Officer Isabella Laura, we are not putting our journalism on platforms that have demonstrated an interest in undermining our credibility

and the public's understanding of our editorial independence, end quote. interesting thing to note here is that in his big interview with the BBC last night, Musk actually said that the state-affiliated media label that BBC and NPR received was going to be changed to publicly funded, not what it actually got changed to, which is government-funded media. So I wonder if he misspoke, if a different decision was made, or if a different outcome may have come from that. But that is the last of your Twitter drama for the day. I'll talk to you again tomorrow.