Month: June 2019

The promise of today’s nascent communications satellite constellations is real: connecting everyone on the globe, no exceptions. But the dark side, or rather bright side, of these satellites threatens to pollute the sky with innumerable points of moving light. Astronomers warn that this may pose a “debilitating threat” if not addressed by regulators or industry.

The International Astronomical Union, a group of over ten thousand astronomers and researchers all over the world, issued a statement this week politely but firmly pointing out the risks of this “new and largely unregulated frontier of space utilisation.”

The problem is that we have graduated from an era where we were launching a satellite every month or so to one where dozens might be launched every week. The competing communications constellations from Starlink, OneWeb, and others will number in the tens of thousands once deployed, outnumbering by far every other satellite in the sky.

This isn’t a question of maybe seeing one little blip out of the corner of your eye — eventually there may be hundreds of these satellites visible at any given time from nearly anywhere on the planet.

Not only that, but these satellites are frequently in relatively low orbits, making them much more visible than distant geosynchronous ones like GPS satellites — not to mention they’re shiny, the IAU writes:

The surfaces of these satellites are often made of highly reflective metal, and reflections from the Sun in the hours after sunset and before sunrise make them appear as slow-moving dots in the night sky.

Are they visible to the naked eye? Depends on a lot of circumstances — the time of day, the position of the craft, the light pollution in your area, and so on. But it’s certainly possible. And astrophotographers and observatories have already noted the problem. You can see the trails of the Starlink satellites in the image at top, and photographer Xplode captured similar trails in this shot — if you look closely you can see there are lots.

High-powered telescopes and sensitive imaging devices are even more susceptible to the issue; the Large Synoptic Survey Telescope team said Starlink was merely a “nuisance,” but acknowledged that for others it may be a far greater threat, and referred to the IAU statement as more representative of the global community.

Naturally there are countermeasures — timing, postprocessing, and other tools for shooters and researchers who don’t want the satellites to interfere. But this will become progressively harder, even impossible for some surveys and exposures, as more satellites are deployed.

SpaceX (which runs Starlink) founder and CEO Elon Musk has tentatively addressed the issue online, promising the team will look into reducing the visibility of the satellites, but that may or may not amount to anything. Any modifications the company makes would be entirely voluntary.

In addition to the visual component, the satellites will be beaming strong radio signals towards the Earth, which can interfere with radio telescopes if the spectrum is not partitioned carefully. This is perhaps a more tractable problem but still must be considered.

This is another case where the industry has long since passed up the regulations that bind it. There’s simply no law against putting thousands of satellites into space — no upper limit on how many, or on how prominent they can be in the sky. The IAU begs that the powers that be look into it:

Satellite constellations can pose a significant or debilitating threat to important existing and future astronomical infrastructures, and we urge their designers and deployers as well as policy-makers to work with the astronomical community in a concerted effort to analyse and understand the impact of satellite constellations. We also urge appropriate agencies to devise a regulatory framework to mitigate or eliminate the detrimental impacts on scientific exploration as soon as practical.

At the speed new regulations are adopted, there will likely be hundreds more satellites in orbit before anything is done, but it’s important to pursue nevertheless. I’ve also asked the American Astronomical Society for their opinion on the topic, and will update this post if I hear back.

As YouTube garners heat for failing to take action on purported hate speech, the company is trying to shift the narrative all while reminding the public that after 14 years it’s still writing the rough drafts of some of its core rules of engagement.

The company announced in a blog post today, that it was expanding the scope of how it would tackle hate speech, now banning language “alleging that a group is superior in order to justify discrimination, segregation or exclusion based on qualities like age, gender, race, caste, religion, sexual orientation or veteran status.”

This blanket ban will also sweep videos related to promoting Nazi ideology as well as content that denies that “well-documented violent events” like the Holocaust or Sandy Hook massacre occurred.

YouTube’s announcement follows a high-visibility show of its inaction after declaring that repeated incidents of being harassment against a Vox writer by right-wing YouTuber Stever Crowder did not violate its policies.

The saga was frustrating, largely because it seemed apparent how YouTube was going to respond long before they did.

As the public looks to YouTube for action tackling hate speech that the company finds bogged in nuance, announcements like this just serve to showcase how unsophisticated the platform’s community guidelines remain. If self-professed Nazis sharing Nazi views has been borderline up to now, how long will it be before the company sees repeated harassment as a cardinal sin?

As the platformized web enters a new age of content moderation, YouTube has proven itself among the most reluctant of powerful actors to make judgment calls.

YouTube is more of a self-contained web than any other platform, almost everything is public and searchable. The scale of the video content uploaded to its servers from public accounts on a daily basis is mind-boggling and presents moderation issues that exist at far greater scales than even Facebook has had to deal with. Luckily, Google remains the global superpower when it comes to AI classification, and yet it seems the company’s largest hinderance to policing is its inability to move deftly in real-time when policy matters arise.

Editor’s note: Drew is a geek who first worked at AOL when he was 16 years old and went on to become a senior writer at TechCrunch. He is now the VP of Communications for venture equity fund Scaleworks.

I have a confession to make. It’s something that I live with daily. It’s not that I’m not proud of it…it’s just that I’m never sure how people will see me after they know my “secret.”

Here goes nothing.

I.

Was.

A.

YouTuber.

Yes, there it is.

Whew, I feel so much better. Or do I? Actually, I don’t. Get ready for a whole lot of “In My Day…” and “It Wasn’t Supposed To Be Like This” because to quote Whoopie Goldberg in Ghost, “Molly, you in danger girl.”

In the beginning

Just to set this up, I’m from Philly. Basically as far away from “Silicon Valley” as you can be in mind and spirit. Before I was a YouTuber, I was a podcaster. A bunch of friends and coworkers of mine wondered out loud, “What if we just recorded ourselves talking about technology? Would anyone listen? Do we even care if anyone listens? No? OK, let’s do it.”

The Best Damn Tech Show, Period. was born.

We went to meetups, podcast tapings, we even got sponsors. Crazy. People kinda listened. We learned about RSS and editing and compression and all kinds of things. It was never a real “side hustle” but just for fun. We had a blast.

Then on February 14, 2005 in the faraway land of San Mateo, CA, a site was born. It was called “YouTube” and it was amazing. Democratization of video. You didn’t have to be famous or important to put anything on it. You could goof off, show a video of your cat doing something stupid and maybe 40 people would watch and you went along on your merry way. It wasn’t an obsession, it was a distraction. Kind of like TV is, but for anyone and everyone.

That was the point.

Google saw potential to the tune of $1.65 billion when they acquired it lock stock and barrel in November 2006. Who knew cat videos could be such a big business? In reality, Google saw ridiculous engagement, mostly on illegal content (which they took on the legal battles without batting an eyelash), that they saw an opportunity to grow it into what it is today…but we’ll get to that later.

Anyways, so here we are, making these fun podcasts and then I asked the group “Hey, what if we put a web cam on us while we talked and uploaded it to this site YouTube. Maybe people would watch us.” It sounded even goofier than podcasting. But we did it.

And it was…something.

But still, we shot our videos, maybe did some editing because we wanted to learn, and then we tossed it up on this site and went along on our merry way.  Sometimes we offended someone, but mostly we just used foul language and talked about Silicon Valley stuff from the perspective of blue collar Philly people.

The years went on and it got more fun. We built a set in a house in Philadelphia and we had actual guests on. (Mostly we just drank and goofed off but it was fun.) The production quality was better, we got more sponsorships, and we kept on doing it for a few years.

Other platforms came along like Blip.tv, and we’d put our stuff on there. But what we were noticing is that people were kinda watching our videos, and then kinda visiting our website and then kinda leaving comments. We had a very small but mighty “Community” and it was cool. And then we went along our merry way, moved all over the country, had families and stopped shooting.

The (de)evolution

The bridge from the “Old” YouTube to the “New” YouTube, in my mind, is when they acquired a company called Next New Networks in 2011.

Former TechCruncher Jason Kincaid’s lede says it all:

Cute kittens and toddlers may be YouTube’s bread and butter, but Google’s video portal needs more than that to encroach on the goliath that is cable TV. But instead of shelling out for the rights to premium content from cable networks, YouTube is hoping it can nudge its existing community toward making high quality videos.

Translation: Spend more time on this stuff and we will make sure you get paid for it.

The game was changed.

From YouTube’s blog in 2011:

In fact, the number of partners making over $1,000 a month is up 300% since the beginning of 2010 and we now have hundreds of partners making six figures a year. But frankly, “hundreds” making a living on YouTube isn’t enough and in 2011 we know we can and should do more to help our partners grow.

Some said that YouTuber PewDiePie makes $12 million a year.

A Year.

Oh yeah, something changed big time in 2011. It’s certainly not fun and games anymore, this is a “business” and when things turn into a business, you lose the fun.

Now I’m not going to get into anymore stats or whatever, but it doesn’t take a rocket scientist to figure out that when millions of people talking about funny cats turns into one person making millions of dollars for their videos, some baaaaad things are going to happen.

Forget YouTube as a “Community” for a second. PewDiePie has a Community. They’re fans. They fight with other Communities to protect their golden child. It gets nasty, and when you zoom out it pretty much looks like a five-year-old arguing on a school blacktop. It’s silly. But when you zoom in, you see something way more nefarious. Those bickering communities drive engagement. A lot of engagement. Engagement is eyeballs, and advertisers pay for eyeballs. They don’t care whose eyeballs and what those eyeballs are watching…they just want to be in front of them.

How does YouTube police this? Well, for the most part, they don’t. Why would they? I say that sarcastically, because of course they should police it. We’re human beings with hearts and souls and feelings. The almighty dollar couldn’t drive most decisions at any company could it? Could it? Of course it does.

Community at scale is an unmitigated disaster

A very smart person who wasn’t always so popular once said “Community Doesn’t Scale” and it threw me into a tailspin. They’re wrong. Look at Facebook, Look At Reddit, Look At YouTube. THEY’RE HUGE!!!”

Clearly, I had missed the point and they were right.

So as this YouTube thing evolved, the happy little coders and engineers and product managers watched how we skipped from cat vid to cat vid, shared them on other sites like MySpace or Fartster or whatever, and built the platform around those tendencies.

Growth hacking. More engagement, more eyeballs, more money, yadda yadda.

So are you surprised that YouTube is now a playground for pedophiles? No..I’m not, because the company that built the growth engine around the early community that just wanted to share a video of their grandfather falling off of his chair while sleeping is now being applied to assholes who call actual other human beings “queer mexicans” for a laugh and views, sans any real humanity behind it.

Duh.

YouTube, with Google’s firepower, has been building this engine since it got its grubby little hands on it in 2006. Twelve Years. It’s taken Twelve Years to become this awful. The Community that YouTube was built around is the Community that is now telling the company that bad things are happening and they might want to take a look at reeling things in before all hell breaks loose and people start dying in large quantities because of the goings on on its site. They call it “personalization” and they say it’s a “tailored experience” and on and on. It’s actually quite laughable how far justification can get you when you don’t really care.

Will YouTube change? HELL NO. Why would they? When you spend that long to perfect being trash, you’re probably pretty proud of it.

Back to “Community”

If you’re a YouTuber user, you’re not a member of its Community. Because if you were, they’d give a shit about you.

Here’s a real definition of “community”: a feeling of fellowship with others, as a result of sharing common attitudes, interests, and goals.

It’s a feeling, not a box. And that feeling isn’t reciprocated from YouTube. How do I know? YouTube’s CEO has proven it:

This may rub some the wrong way…but that’s a flat out lie.

Even today’s chest-beating over removing some supremacist and hateful content comes off as extremely weak and self-serving:

“Thanks to these investments, videos that violate our policies are removed faster than ever and users are seeing less borderline content and harmful misinformation.”

Do you want a cookie?

Look, just because we all like to watch videos, we’re a member of a “Community”? Nope, doesn’t work like that. You’re a cog in a wheel, you’re the eyeballs, you’re not a part of something big and great. Your feelings and well-being simply don’t matter. Sucks, but as my dad used to say “dems da breaks.”

Community doesn’t scale, because by nature…it can’t. YouTube may grow in size, but it will grow without the people that they originally built the platform for and around. Shedding annoying skin like a snake. It will evolve, but for the people who use it today for whatever they use it for today that makes more money for YouTube. Make no mistake, their best interests won’t be kept in mind either, just like the original crowd’s wasn’t and weren’t.

The Steven Crowder’s of the world will move out and be replaced by some new jackhole group of people that likes to fish fight each other to the soundtrack of the Terminator or something. Whatever it is, it’ll be weird. And yes, it’ll come with all new problems. And no, YouTube won’t do anything to “fix” them.

The company just doesn’t care.

So yep, things have changed. I’m an “old” but I’m not asking you to get off my lawn. I’m going to get off yours. I had fun making goofy videos and learning how to make them. Hearing from people who liked our stuff was neat. But we moved on. And with YouTube as a whole, I’ll move on. Because it’s not mine anymore.

Maybe I’ll find a new one.

Have fun, and be safe.

MIT researchers have created a new autonomous robot boat prototype – which they have named ‘roboats’ to my everlasting glee – that can target and combine with one another Voltron-style to create new structures. Said structures could be bigger boats, but MIT is thinking a bit more creatively – it envisions a fleet of these being able to join up to form on-demand urban infrastructure, including stages for concerts, walking bridges or even entire outdoor markets.

The roboats would of course be able to act as autonomous water taxis and ferries, which could be particular useful in a setting like Amsterdam, which is why MIT teamed up with Amsterdam’s Institute for Advanced Metropolitan Solutions on this joint. Equipped with sensors, sub-aquatic thrusters, GPS, cameras and tiny computer brains, the roboats can currently follow a pre-determined path, but testing on newer 3D-printed prototypes introduced a level of autonomy that can accomplish a lot more.

New tests focused on a a custom latching system, which a very high degree of precision, that can connect to specific points with millimetre accuracy, using a trial and error algorithm-based autonomous programming to make sure they connect to their target correctly. The initial use case in Amsterdam that MIT identified is overnight garbage collection, where these could act as mini barges working the canal to quickly and easily clear refuse left out by residents and store owners.

Longer-term, the vision is to see what kind of additional configurations might be possible, including larger platforms that can support people on board, and “tentacle-like rubber grippers that tighten around the pin — like a squid grasping its prey” to improve the latching mechanism in a way inspired by a somewhat terrifying visual.

At its re:Mars conference, Amazon today announced that it is working on making interacting with its Alexa personal assistant more natural by enabling more fluid conversations that can move from topic to topic — and without having to constantly say ‘Alexa.’

At re:Mars, the company showed a brief demo of how this would work to order movie tickets, for example, taking you from asking ‘Alexa, what movies are playing nearby?’ to actually selecting the movie, buying the tickets and making a restaurant reservation nearby — and then watching the trailer and ordering an Uber.

In many ways, this was a demo that I would have expected to see from Google at I/O, but Amazon has clearly stepped up its Alexa game in recent months.

The way the company is doing this is by relying on a new dialogue system that can predict next actions and easily switch between different Alexa skills. “Now, we have advanced our machine learning capabilities such that Alexa can predict customer’s true goal from the direction of the dialogue, and proactively enable the conversation flow across skills,” the company explained.

This new experience, which Amazon demoed on an Alexa Show, with the appropriate visual responses, will go live to users in the coming months.

Over the last few months, the company also announced today, Alexa also became 20 percent more accurate in understanding your requests.

In addition, developers will be able to make use of some of these technologies that Amazon is using for this new dialogue tool. This new tool, Alexa Conversations, allows developers to build similar flows. Traditionally, this involved writing lots of code, but Alexa Conversations can bring this down by about a third. The developer simply has to declare a set of actions and a few example interactions. Then, the service will run an automated dialog simulator so that the developers don’t actually have to think of all of the different ways that a customer will interact with their skills. Over time, it will also learn from how real-world users interact with the system.

The Daily Crunch is TechCrunch’s roundup of our biggest and most important stories. If you’d like to get this delivered to your inbox every day at around 9am Pacific, you can subscribe here.

1. Connected bike and treadmill-maker Peloton files confidentially for IPO

While Peloton made the announcement of its filing in a press release, it did not disclose the terms of its IPO.

The company, which has been valued at $4 billion, paved the way for a variety of home fitness companies. It expanded beyond bikes earlier this year with the release of a connected treadmill with similar services and content.

2. SEC expands its war on cryptocurrency companies with a lawsuit against Kik

In the lawsuit, the SEC claims that Kik conducted an illegal $100 million offering of digital tokens by selling the tokens to U.S. investors without registering their offer and sale as required under U.S. law.

3. YouTube says homophobic taunts don’t violate its policies

YouTube has told a reporter at Vox that the comments made by a prominent conservative channel (describing the reporter as as “lispy queer” and “the gay Latino host at Vox”) do not violate its policies.

4. Apple looks to recharge its broader app ecosystem at WWDC 2019

At WWDC this week, Apple launched a new user interface framework, called SwiftUI. It’s designed to help developers build a full-featured user interface with smooth animations using simple, declarative code.

5. 7.7 million LabCorp records stolen in same hack affecting Quest

LabCorp is the latest laboratory testing giant to confirm this week that it’s affected by the same third-party data breach.

6. SentinelOne raises $120M for its fully autonomous, AI-based endpoint security solution

The startup provides real-time endpoint protection on laptops, phones, containers, cloud services and most recently IoT devices on a network through a completely autonomous, AI-based platform.

7. KLM Airlines wants to help build a more efficient jet with in-wing seating

You may also notice a familiar byline here: Darrell Etherington is back! And yes, he says he’ll help out with the Daily Crunch, though you’re still stuck with me on most days.

Nine months ago, the once high-flying venture capital fund Social Capital made the bold decision to stop accepting outside capital and operate as a family office, in essence.

The co-founder of the outfit, brazen billionaire and early Facebook executive Chamath Palihapitiya, pledged to upend his investment strategy and make fewer but much larger investments as a means to improve his returns. Naturally, a near-complete exodus of Social Capital’s venture capitalists followed.

Today, the firm’s three founders, Palihapitiya, Mamoon Hamid and Ted Maidenberg, have gone their separate ways. Palihapitiya is rewriting the Social Capital playbook, Hamid is busy reinvigorating Kleiner Perkins and Maidenberg is building on top of the data-driven strategy and proprietary software dubbed “Magic 8-Ball” he built at Social Capital, with a new firm called Tribe Capital.

Quietly, Tribe Capital’s co-founders, Maidenberg and former Social Capital partners Arjun Sethi and Jonathan Hsu, have deployed millions of dollars in Social Capital portfolio companies like Slack and Carta, hired several former Social Capital employees and flexed a data-first approach that looks pretty damn familiar.  

Data or bust

Social Capital began laying the foundation for a data-driven approach to investing years ago. Now, Tribe Capital is doubling down.

From its founding in 2011, Social Capital established itself as a contrarian fund out to “fix capitalism.” Its strategy and reputation as an up-and-comer unafraid of new tricks earned it stakes in Slack, SurveyMonkey, Box, Bust and many other admirable upstarts.

As the firm matured, its partners experimented. In 2016, its early-stage investment team made the daring choice to rely on data rather than gut-feel alone to make its investment decisions, confronting a timeworn ideology that the best VCs have a special skill-set that enables them to spot future unicorns.

Using an operating system for early-stage investing dubbed “capital-as-a-service” and the growth and data analysis tool Magic 8-Ball — a sort of QuickBooks for startup data — Social Capital forwent the traditional pitch process and rapidly evaluated thousands of companies on the basis of metrics and achievements alone.

Palihapitiya, Maidenberg, Hamid and the other members of the partnership were on a mission to do venture the right way. Until they weren’t.

“I found us incrementally drifting away from our core mission, and our strategy was increasingly that of a traditional investment firm,” Palihapitiya wrote last year. “It became harder to take the risks we took in 2011 and it became easier to play the same game as every other VC.”

At its peak, Social Capital employed a team of 80. Once Palihapitiya confirmed his intent to transition the firm away from venture, the team began to shrink, fast. Today, the firm employs 30, including partners Ray Ko, Andy Artz and Jay Zaveri, as well as principal Alex Danco. One-third of that number were hired after the big pivot.

The Social Capital diaspora 

Social Capital’s former investors have since identified their second acts.

In the last year, Sakya Duvvuru, a former partner, founded Nellore Capital Management, and Carl Anderson, another former partner, started Marcho Partners.

Tony Bates joined Genesys as its CEO, Mike Ghaffary accepted a general partner role at Canvas Ventures, Ashley Carroll is consulting full-time, Kristen Spohn says she is still exploring opportunities, Adam Nelson joined South Park Commons as a venture partner and Tejinder Gill joined Collaborative Fund as a principal.

Hamid, for his part, resolved to re-establish Kleiner Perkins’ once-stellar reputation.

“Kleiner Perkins was a firm that was in desperate need of a change of its own,” Hamid tells TechCrunch. “It was a unique opportunity and I was about to turn 40. I thought, there is one thing I wanted to do in my career that I hadn’t done before and that was to turn around one of the best venture firms of all time.”

Hamid’s August 2017 departure from Social Capital represented the beginning of the end of the partnership. Though Hamid, a co-founder and leading dealmaker, asserts turmoil at the firm began after his exit. 

Nine months after Hamid made the call to move on, Arjun Sethi, who once led Social Capital’s early-stage investment team, made the same call as did Maidenberg and Hsu. Simultaneously, growth equity chief Tony Bates and vice chairman Marc Mezvinsky were said to be departing.

The mass exodus continued, culminating in Palihapitiya’s final declaration: Social Capital was finished with venture capital.

‘Magic 8-Ball’ — reborn

Maidenberg, Sethi and Hsu built Tribe Capital in the image of Social Capital. With similar DNA, the three men are attempting to upgrade an early-stage investment strategy they not only created, but nearly perfected.

“Those guys did a very good job working for me,” Palihapitiya tells TechCrunch. “I’m super proud to see them launch their own venture fund. It was a really important, defining experience for me; I hope they have the same level of success, if not more.”

But where Social Capital was mission-driven, regularly backing healthcare and education businesses, Tribe Capital makes no such claim. And where Social Capital leaned on data to inform its investment thesis, Tribe is putting its full weight into it.

“We are believers that it’s hard to do a lot of things well, so we wanted to focus on one thing we are good at: early-stage venture with the approach of recognizing early-stage product-market fit,” Hsu tells TechCrunch. “At Social Capital we did that, but we did 30 other things, too.”

In total, seven former Social Capital investors and employees are working on Tribe. Georgia Kinne, a former Social Capital executive assistant, leads operations. Two former Social Capital data scientists, Jake Ellowitz and Brendan Moore, joined Tribe in the same role. And Alexander Chee, Social Capital’s former head of product development, is on board as an entrepreneur-in-residence.

Tribe won’t say how much capital they have raised yet or how exactly their three funds are structured, aside from confirming that only one is operating as a traditional venture fund. Paperwork filed with the U.S. Securities and Exchange Commission in late April, however, confirms a $150 million target for the debut venture effort. 

It’s been a year since Tribe began investing. In that time, it’s put money in Slack, Front, Cover and SFox. Most recently, it participated in Carta’s $300 million Series E, which valued the business at $1.7 billion. All of these companies were previously backed by Social Capital.

Tribe is making deals of all shapes and sizes across industries, with a particular focus on enterprise, fintech and SaaS startups. In addition to deploying heftier sums to late-stage businesses like Slack, Tribe has made 10 seed bets of roughly $25,000 each, leveraging its data platform to make investment calls.

“The income statement and balance sheet are the lingua franca for an established company to communicate the financial health of its business,” Hsu writes. “These accounting concepts are often unhelpful when inspecting an unprofitable early-stage company. For a startup, what’s needed is a common quantitative language for what matters, namely, a quantitative framework for assessing product-market fit.”

Tribe’s quantitative framework is called Magic 8-Ball, a diligence tool for potential investments created by Maidenberg and Hsu during their Social Capital tenure. The tool measures product-market fit, growth trajectory and more of early-stage businesses, where, as Hsu mentions, financial data may be lacking.

“We use data like accountants; it’s not a magical AI machine,” Hsu said. “If other firms want to copy, by all means, they can try. We aren’t here to be antagonistic, we are here to be partners to founders and other investors.”

So far, Magic 8-Ball has poured through data provided by some 200 companies, with plans to hit 1,000 per year. In total, Tribe has deployed $100 million.

Tribe’s 8-Ball tool is said to be much more complex than the earlier model, according to a source with knowledge of the platform. It’s like when Yahoo engineers Jan Koum and Brian Acton left the search and email giant to build something even better, the source, who asked not to be named, said. That business became the messaging powerhouse WhatsApp.

Hamid, who’s not affiliated with Tribe but aware of their investment strategy, made a similar comparison.

“It’s like if you’re an engineer at Cisco working on WebEx,” Hamid tells TechCrunch. “You’re a great engineer but you can do better, you can [do your own] company. Guess what? That’s Zoom. That’s Eric Yuan . And Zoom is worth $20 billion and WebEx was worth $3 billion. That’s pretty. That’s the story of Silicon Valley. That’s creative disruption.”

Hamid, however, was careful to point out the differences between Social Capital and Tribe. The DNA may be similar but they aren’t identical.

Social Capital represented a new kind of venture firm in favor of creative disruption. Tribe Capital represents a second go, a sort of Social Capital 2.0 sans Chamath Palihapitiya.

Bogged down by the conflict surrounding its leader’s flair for controversy, Social Capital wasn’t set up to succeed. The Magic-8 Ball, on the other hand, may be just right.

“Why did we get back together instead of going elsewhere? That is a reasonable question,” Hsu said. “We had good job offers but we had a viewpoint of the world that we wanted to keep working on together.”

After a relatively quiet show last year, Computex picked up the pace this year, with dueling chip launches by rivals AMD and Intel and a slew of laptop releases from Asus, Qualcomm, Nvidia, Lenovo and other companies.

Founded in 1981, the trade show, which took place last week from May 28 to June 1, is one of the ICT industry’s largest gatherings of OEMs and ODMs. In recent years, the show’s purview has widened, thanks to efforts by its organizers, the Taiwan External Trade Development Council and Taipei Computer Association, to attract two groups: high-end computer customers, such as hardcore gamers, and startups looking for investors and business partners. This makes for a larger, more diverse and livelier show. Computex’s organizers said this year’s event attracted 42,000 international visitors, a new record.

Though the worldwide PC market continues to see slow growth, demand for high-performance computers is still being driven by gamers and the popularity of esports and live-streaming sites like Twitch. Computex, with its large, elaborate booths run by brands like Asus’ Republic of Gaming, is a popular destination for many gamers (the show is open to the public, with tickets costing NTD $200, or about $6.40), and began hosting esport competitions a few years ago.

The timing of the show, formally known as the Taipei International Information Technology Show, at the end of May or beginning of June each year, also gives companies a chance to debut products they teased at CES or preview releases for other shows later in the year, including E3 and IFA.

One difference between Computex now and ten (or maybe even just five) years ago is that the increasing accessibility of high-end PCs means many customers keep a close eye on major announcements by companies like AMD, Intel and Nvidia, not only to see when more powerful processors will be available but also because of potential pricing wars. For example, many gamers hope competition from new graphic processor units from AMD will force Nvidia to bring down prices on its popular but expensive GPUs.

The Battle of the Chips

The biggest news at this year’s Computex was the intense rivalry between AMD and Intel, whose keynote presentations came after a very different twelve months for the two competitors.

For the first time, Amazon today showed off its newest fully electric delivery drone at its first re:Mars conference in Las Vegas. Chances are, it neither looks nor flies like what you’d expect from a drone. It’s an ingenious hexagonal hybrid design, though, that has very few moving parts and uses the shroud that protects its blades as its wings when it transitions from vertical, helicopter-like flight at takeoff to its airplane-like mode.

These drones, Amazon says, will start making deliveries in the coming months, though it’s not yet clear where exactly that will happen.

What’s maybe even more important, though, is that the drone is chock-full of sensors and a suite of compute modules that run a variety of machine learning models to keep the drone safe. Today’s announcement marks the first time Amazon is publicly talking about those sensors, which it designed in-house, and how the drone’s autonomous flight systems maneuver it to its landing spot. The focus here was on building a drone that is as safe as possible and able to be independently safe. Even when it’s not connected to a network and it encounters a new situation, it’ll be able to react appropriately and safely.

When you see it fly in airplane mode, it looks a little bit like a TIE fighter, where the core holds all the sensors and navigation technology, as well as the package. The new drone can fly up to 15 miles and carry packages that weigh up to five pounds.

This new design is quite a departure from earlier models. I got a chance to see it ahead of today’s announcement and I admit that I expected a far more conventional design — more like a refined version of the last, almost sled-like, design.

Besides the cool factor of the drone, though, which is probably a bit larger than you may expect, what Amazon is really emphasizing this week is the sensor suite and safety features it developed for the drone.

Ahead of today’s announcement, I sat down with Gur Kimchi, Amazon’s VP for its Prime Air program, to talk about the progress the company has made in recent years and what makes this new drone special.

“Our sense and avoid technology is what makes the drone independently safe,” he told me. “I say independently safe because that’s in contrast to other approaches where some of the safety features are off the aircraft. In our case, they are on the aircraft.”

Kimchi also stressed that Amazon designed virtually all of the drone’s software and hardware stack in-house. “We control the aircraft technologies from the raw materials to the hardware, to software, to the structures, to the factory to the supply chain and eventually to the delivery,” he said. “And finally the aircraft itself has controls and capabilities to react to the world that are unique.”

What’s clear is that the team tried to keep the actual flight surfaces as simple as possible. There are four traditional airplane control surfaces and six rotors. That’s it. The autopilot, which evaluates all of the sensor data and which Amazon also developed in-house, gives the drone six degrees of freedom to maneuver to its destination. The angled box at the center of the drone, which houses most of the drone’s smarts and the package it delivers, doesn’t pivot. It sits rigidly within the aircraft.

It’s unclear how loud the drone will be. Kimchi would only say that it’s well within established safety standards and that the profile of the noise also matters. He likened it to the difference between hearing a dentist’s drill and classical music. Either way, though, the drone is likely loud enough that it’s hard to miss when it approaches your backyard.

To see what’s happening around it, the new drone uses a number of sensors and machine learning models — all running independently — that constantly monitor the drone’s flight envelope (which, thanks to its unique shape and controls, is far more flexible than that of a regular drone) and environment. These include regular camera images and infrared cameras to get a view of its surroundings. There are multiple sensors on all sides of the aircraft so that it can spot things that are far away, like an oncoming aircraft, as well as objects that are close, when the drone is landing, for example.

The drone also uses various machine learning models to, for example, detect other air traffic around it and react accordingly, or to detect people in the landing zone or to see a line over it (which is a really hard problem to solve, given that lines tend to be rather hard to detect). To do this, the team uses photogrammetrical models, segmentation models and neural networks. “We probably have the state of the art algorithms in all of these domains,” Kimchi argued.

Whenever the drone detects an object or a person in the landing zone, it obviously aborts — or at least delays — the delivery attempt.

“The most important thing the aircraft can do is make the correct safe decision when it’s exposed to an event that isn’t in the planning — that it has never been programmed for,” Kimchi said.

The team also uses a technique known as Visual Simultaneous Localization and Mapping (VSLAM), which helps the drone build a map of its current environment, even when it doesn’t have any other previous information about a location or any GPS information.

“That combination of perception and algorithmic diversity is what we think makes our system uniquely safe,” said Kimchi. As the drone makes its way to the delivery location or back to the warehouse, all of the sensors and algorithms always have to be in agreement. When one fails or detects an issue, the drone will abort the mission. “Every part of the system has to agree that it’s okay to proceed,” Kimchi said.

What Kimchi stressed throughout our conversation is that Amazon’s approach goes beyond redundancy, which is a pretty obvious concept in aviation and involves having multiple instances of the same hardware on board. Kimchi argues that having a diversity of sensors that are completely independent of each other is also important. The drone only has one angle of attack sensor, for example, but it also has a number of other ways to measure the same value.

Amazon isn’t quite ready to delve into all the details of what the actual on-board hardware looks like, though. Kimchi did tell me that the system uses more than one operating system and CPU architecture, though.

It’s the integration of all of those sensors, AI smarts and the actual design of the drone that makes the whole unit work. At some point, though, things will go wrong. The drone can easily handle a rotor that stops working, which is pretty standard these days. In some circumstances, it can even handle two failed units. And unlike most other drones, it can glide if necessary, just like any other airplane. But when it needs to find a place to land, its AI smarts kick in and the drone will try to find a safe place to land, away from people and objects — and it has to do so without having any prior knowledge of its surroundings.

To get to this point, the team actually used an AI system to evaluate more than 50,000 different configurations. Just the computational fluid dynamics simulations took up 15 million hours of AWS compute time (it’s good to own a large cloud when you want to build a novel, highly optimized drone, it seems). The team also ran millions of simulations, of course, with all of the sensors, and looked at all of the possible positions and sensor ranges — and even different lenses for the cameras — to find an optimal solution. “The optimization is what is the right, diverse set of sensors and how they are configured on the aircraft,” Kimchi noted. “You always have both redundancy and diversity, both from the physical domain — sonar versus photons — and the algorithmic domain.”

The team also ran thousands of hardware-in-the-loop simulations where all the flight services are actuating and all the sensors are perceiving the simulated environment. Here, too, Kimchi wasn’t quite ready to give away the secret sauce the team uses to make that work.

And the team obviously tested the drones in the real world to validate its models. “The analytical models, the computational models are very rich and are very deep, but they are not calibrated against the real world. The real world is the ultimate random event generator,” he said.

It remains to be seen where the new drone will make its first deliveries. That’s a secret Amazon also isn’t quite ready to reveal yet. That will happen within the next few months, though. Amazon started drone deliveries in England a while back, so that’s an obvious choice, but there’s no reason the company could opt for another country as well. The U.S. seems like an unlikely candidate, given that the regulations there are still in flux, but maybe that’s a problem that will be solved by then, too. Either way, what once looked like a bit of a Black Friday stunt may just land in your backyard sooner than you think.

On its face, Amazon’s first Re:MARS conference is all about far out, world changing ideas, but the company is still very much a retailer at heart. Fitting then, that one of the first big announcements from this morning’s event is all about using artificial intelligence to help people better shop for close.

Amazon’s been talking about similar initiatives for a while, but StyleSnap looks to actually be coming soon via the Alexa iOS app (though the actual timeframe is still TBD).

Amazon’s Consumer Worldwide CEO Jeff Wilke introduced the feature today, telling the crowd, “When a customer uploads an image, we use deep learning for object detection to identify the various apparel items in the image and categorize them into classes like dresses or shirts. We then find the most similar items that are available on Amazon.”

The feature will be accessible by clicking the camera icon in the corner of the Alexa app. Users take a photo or upload a screenshot of a look they like, and Amazon will offer up suggestions, that factor in things like price, reviews and brands.

The company’s got a blog post detailing some of the steps taken in order to provide the service. It’s a series of complex asks for what seemingly amounts to a simple task for the human brain.

“To have neural networks identify a greater number of classes, we can stack a greater number of layers on top of each other,” the company writes. “The first few layers typically learn concepts such as edges and colors, while the middle layers identify patterns such as “floral” or “denim”. After having passed through all of the layers, the algorithm can accurately identify concepts like fit and outfit style in an image.”

Amazon has already begun to implement AI for other shopping applications, including Go and Whole Foods.