Tech Brew Ride Home - Wed. 12/26 - Airbnb and Slack Mull Untraditional IPOs

Episode Date: December 26, 2018

How the US government shutdown affects cybersecurity, how the Open Government Data Act is possibly good tech governance, Airbnb and Slack are considering non-traditional IPOs and the state of AI research at the end of 2018. How a government shutdown affects America’s cybersecurity workforce (FifthDomain) In a huge win for open data, Congress passes the Open, Public, Electronic, and Necessary Government Data Act (BoingBoing) Wall Street Quietly Shelves Its Bitcoin Dreams (Bloomberg) Layoffs Underway Amid ‘Adjustments,’ Bitcoin Miner Bitmain Confirms (CoinDesk) HQ Trivia launches HQ Words as reinstalled CEO seeks a game-changer (TechCrunch) Airbnb and Slack are considering untraditional IPOs that box out bankers like Spotify did (Recode) Geoffrey Hinton and Demis Hassabis: AGI is nowhere close to being a reality (VentureBeat) 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 TechMeme ride home for Wednesday, December 26, 2018. I'm Brian McCullough. Today, how the U.S. government shutdown is affecting cybersecurity,

how the Open Government Data Act is possibly good tech governance, Airbnb and Slack are considering non-traditional IPOs, and the state of AI research at the end of 2018. Here's what you missed today in the world of tech. Well, not only are a lot of folks in the U.S. not doing a ton of work this week, There's also the little matter of the government shutdown that hit right before the holidays began. As you may or may not be aware, there are certain areas within the U.S. government that are immune to shutdowns, certain jobs that are still on the job, the so-called essential employees.

For example, when we traveled during the shutdown on Sunday, TSA workers, we're still checking our suitcases at the airport. But guess what? Among the government employees not considered a, and thus furloughed for the duration of the shutdown? Well, the U.S.'s cybersecurity workforce. The National Institute of Standards and Technology, which sets cybersecurity regulations and, in fact, was scheduled to release a major report in the coming weeks, has seen 85% of its workforce furloughed. The Director of National Intelligence's Analysis and Operations workforce has also reportedly

seen a 60% reduction in its active workforce. But quoting from a fifth domain report from last week before the shutdown happened, quote, it appears that the Department of Homeland Security's new cybersecurity and infrastructure security agency created just last month is among the most protected in the event of a government shutdown. The agency would only have 45% of its workforce furloughed with 2008 employees exempt. Spokespeople from the Department of the Department of Homeland Security did not respond to repeated requests for comment regarding how the shutdown would affect their agency. My hope would be that anyone at DHS who is working on cyber, those would be viewed as critical positions and would continue working, Senator Mark Warner of Virginia

told Fifth Domain, end quote. Yes, I would hope so too. Wasn't it just last week that we were hearing those rumblings and ramping up of accusations about major Chinese government hacking attempts of U.S. government and technology industry systems. It's unlikely, of course, that those efforts will be suddenly put on pause, and that's not even taking into account run of the mill hackers and nefarious actors. Maybe it's time to update the definitions of essential workers in the 21st century. We know that crime never sleeps, and just because the cops stand down, that doesn't mean the bad actors do either.

Having said that, I think it's also important to give credit when the government gets something right, when it comes to tech policy, Congress has passed the Open Government Data Act, which requires public information to be open to the public by default in machine readable formats. In essence, this brings us into the 21st century sort of thing again, publishing all non-sensitive government data, legislation, etc., in a format that is easily legible, searchable, and disseminatable, in a format easily readable from a smartphone or a laptop. say. In other words, in a modern democratic society, the machinations of government should be published in a way that allows the citizenry to be easily informed. The bill also stipulated that

federal agencies should use and cite publicly available evidence when they make policy. Quoting Corey Doctor-O in Boing Boing, the bill, which was exemplary in its original form, was somewhat neutered before it passed, carving out government agencies, save those defined by the CFO Act. and data that does not concern monetary policy, meaning that Treasury is exempted from this presumption of openness. Despite the carve-outs, this is still a big deal, and it's hard to say what makes it a bigger deal, the open data or the mandate for evidence-based policy. The administration's expert agencies like the FCC are already required to work on the basis of evidence, which is why it's such a big deal that Ajit Pai ignored expert opinion in formulating the net neutrality-killing FCC,

order, making the whole order vulnerable to legal challenge. And of course, open data and evidence-based policy go hand in hand. Open data gives you the factual basis to evaluate the policies, end quote. As Rand Fishkin tweeted, quote, I'm often critical of what the current U.S. Congress is doing, but this is a rare positive move that deserves recognition. If you weren't aware, over the weekend, HQ Tribia officially launched HQ Words, that Wheel of Fortune like word game that I I have told you about before that hopefully will be a case of lightning striking twice for the company in terms of virtual game show style apps. Co-founder Russ Usapov told TechCrunch, quote, Intermedia Labs introduced the world to a category-defining product, HQ trivia. Once again,

with HQ Words, Intermedia Labs is poised to captivate the world with a revolutionary experience that will bring people together in new ways around live mobile video. HQ Words is the most interactive experience we've ever made, end quote. HQ words can be found inside the existing HQ trivia app, and it works on a similar premise. There's a 25-second timer. There are words and phrases hidden so that you have to slowly guess the identity of the word. And if you choose three wrong letters or fail to fill in the puzzle in time, you're out. And you're competing against the crowd in a last player standing sort of paradigm, just like HQ trivia. If you make it through 10 rounds, you get a share of the cash prize,

and those who solve the puzzle's fastest get a larger cut of the kitty. And if this is Wheel of Fortune to HQ trivia's Jeopardy, then the power hour programming block is carried over as well. Words goes live daily at 6.30 p.m. Pacific time after trivia's 6 o'clock p.m. game. Worth noting that Yuzapov has officially been named CEO of overall company Intermedia Labs after the tragic death of fellow co-founder and former CEO, Colin Kroll. In a tribute to Kroll, Yuzapov wrote, quote, Colin and I shared many incredible life moments over the past seven years. We embarked on an incredible journey, co-founding two breakthrough companies together, and the lessons we learned at Vine and HQ will continue to have a big impact on me. Like many relationships, we've also had our challenges, but it was during these challenging times

that Collins' kind soul and big heart would truly shine. The driving force behind his innovations was the positive impact they would have on people and the world. Collins' innovations and inventions have changed many people's lives for the better and will continue to impact the world for years to come, end quote. We spoke a lot last week about the flood of big name tech IPOs that we are expecting over the next year and many of them in the next three to six months. Recode is reporting that while Uber and Lyft are expected to pursue traditional IPOs underwritten by Wall Street, Airbnb and Slack are considering non-traditional IPOs, the sort that, to a degree, box out big bankers, much in the way that Spotify's listing did.

If you'll recall, Spotify went public via a so-called direct listing. At the time, some were wondering if this would be a trend-setting event, which, with other going public companies tempted to follow suit. Well, according to Recode, Airbnb is especially kicking the tires of a direct listing, quote. The conversations have been serious enough that CEO of Airbnb Brian Chesky has consulted in recent months with the CEO of Spotify, Daniel Eck, about how Airbnb could possibly pursue its own similar listing, the people say. Airbnb declined to comment. Chesky is said to be interested in making Airbnb's public offering more than merely a fine,

event and is drawn to any approach that makes his IPO less traditional, like perhaps granting Airbnb hosts stock, end quote. As for Slack, quote, Slack's CEO Stuart Butterfield is said to be curious about the idea, though he has not yet made any final decisions, according to people familiar with his thinking. Slack's offering isn't expected until the second half of 2019. What a difference a year makes. The highs of the big. crypto run-up were exactly a year ago right now, and I'm not pointing that out to pile on, and I'm not bringing up these stories to do the same, but instead continuing to cover how the widespread crypto crash over the last year is having a real impact and manifesting itself

in several ways. Bloomberg is reporting that the highly publicized crypto efforts of Wall Street, by the likes of Goldman Sachs, Morgan Stanley, Barclays, and others have, shall we say, stalled a bit. among weak demand from institutional clients for crypto assets, as well as perhaps the latent skepticism that Wall Street already had towards crypto, rising once again back to the four. Take the example of Goldman, among the first on the street to clear Bitcoin futures, quoting Bloomberg, the bank has yet to offer trading of crypto and has gained little traction for its NDF product,

having signed up just 20 clients, according to people familiar with the matter. Justin Schmidt, who was hired to head its digital asset business, said at an industry conference last month that regulators are limiting what he can do. Still, Goldman plans to add a digital assets specialist to its prime brokerage division, the person said. Morgan Stanley, which hired Andrew Peel as its head of digital assets earlier in the year, has been technically prepared to offer swaps tracking Bitcoin futures since at least September, yet thus far has not traded a single contract, according to a person familiar with the matter. A person with knowledge of the business said in September the contracts would be launched once there is proven institutional client demand, end quote. Similar stories at City, at Barclays, as Bloomberg describes that a lot of these projects are in, quote, limbo waiting to see what happens. Quote, squeamish from the start about pursuing profits in one of the darker corners of finance. Established firms this year slowed their already halting efforts to make a business out of Bitcoin Mania.

while none has thrown in the towel and some continued to develop a trading infrastructure, most flinched as the value of virtual coins collapsed, end quote. Of course, as I told Chris Dixon on the A16Z podcast last week, it was right when the whole world decided that the internet was a fad and everyone shuddered their internet divisions that things like Google and Facebook slowly started to take over the world. More on that, by the way, at the end of the show. But also, real quick, across the Pacific, I swear it was not that long ago, even less than a year ago, August maybe, that I was telling you about Bitmain, planning an IPO that could have been the largest tech IPO in history. Well, in this case, what a difference three months can make.

Because CoinDesk is reporting that employees at the Crypto Miner Behemoth have anonymously confirmed layoffs at BitMain, saying some departments of the company are going to disappear entirely. Quoting CoinDesk, since it is still an ongoing process, the source said it is unclear how many people have been impacted so far and was not able to verify the claim of 50% layoffs. But the whole thing certainly couldn't be handled in just one day given the total number could be large, the source said, adding, this is an operational adjustment. Some projects will be entirely gone, so it's hard to calculate a precise percentage at this stage, end quote. Finally, today, I'll leave you with a legit long read that you can peruse at your leisure, but leave you with the headline quotes first. I feel like this year was the year that the promise and concerns of AI really broke through to the mainstream. Maybe my biggest bet will be when we will actually see AI that can pass the Turing test,

AI that can do the thinking that a, say, six-year-old human brain can do. It's not just that this year people started to think seriously about AI beating humans at games. That's been going on for many years. Or the possibility of autonomous vehicles right around the corner or the fear of jobs disappearing, although maybe that's what really has come to the forefront lately, as well as countries worrying increasingly that falling behind an AI means falling behind economically. McKinsey, for example, predicts that in the U.S. alone, AI could capture 25% of all economic benefits in the next decade or so. It's also, though, about the idea that true AI,

otherwise known as artificial general intelligence or AGI, itself is on the horizon. Well, Venture Beat quotes extensively from two giants in the AI field to suggest that, as with my other bets, perhaps we're not quite there yet. Demise Hasebis is the co-founder of Deep Mind, and when it comes to AGII or something, systems that could perform any intellectual task of cognition equal to what a human could do, quote, they're still much farther to go. Games or board games are quite easy in some ways because the transition model between states is very well specified and easy to learn. Real world 3D environments and the real world itself is much more tricky to figure out,

but it's important if you want to do planning. We don't have systems that can transfer in an efficient way knowledge they have from one domain to the next. I think you need things like concepts or extractions to do that. Building models against games is relatively easy because it's easy to go from one step to another, but we would like to be able to imbue systems with generative model capabilities, which would make it easier to do planning in those environments, end quote. Jeffrey Hinton of Google's Deep Brain Learning Team addressed the concerns that have arisen this year that AI systems could end up being biased

because they are trained incorporating the biases that might be inherent in existence, data or inherent in the humans training the systems themselves. But Hinton is actually quite optimistic about that wrinkle. Quote, anything that learns from data is going to learn all the biases in the data. The good news is that if you can model biases in the data, you can counteract them pretty effectively. There's all sorts of ways of doing that. Conversely, he says, quote, if you have people doing the jobs, you can try and model their biases. Telling them not to be biased doesn't quite work like subtracting the biases. So I think it'll be much easier in a machine learning system to deal with bias, end quote.

Again, this one is a legit long read that takes the temperature of the state of the art in AI as we enter 2019, discussing recent breakthroughs like deep neural networks that are modeled on the human brain, and what are known as Transformers, a new type of neural architecture that requires less computational reps, as it were, to train. Check it out. Last link in the show notes.

My thanks to Chris Dixon and everyone at Andreessen Horowitz for having me on the latest episode of the A16Z podcast to talk about my book, how the internet happened. That episode, it's actually the best conversation I've had promoting the book. Not only did we get into the history,

of tech over the last quarter century, but also Chris and I talked a lot about the underlying structural and cultural and economic forces that have gotten tech to where it is today as an industry. And we talk at length about where tech is at this very moment, building off of things that we've actually talked about on this very podcast. One of the reasons actually why I love doing this show for you all every day is that it kind of helps me develop and flesh out ideas about tech trends that are immediate and surfacing in almost real time. In a lot of ways, I feel like doing this podcast and perhaps I hope listening to this podcast every day allows a degree of keeping your finger on the pulse of what's happening right now in tech that I think is unparalleled. Anyway, check that out.

The most recent A16Z podcast episode. Talk to y'all tomorrow.