A reminder for new readers. That Was The Week collects the best writing on critical issues in tech, startups, and venture capital. I select the articles because they are of interest. The selections often include things I disagree with. The articles are only snippets. Click on the headline to go to the original. I express my point of view in my editorial and the weekly video.
Content this week from @stuwoo @mvpeers @JuliaAngwin @chloexiang @vice @neuwaves @jason_keobler @tayhatmaker @pkedrosky @blurcon @billgates @theo_wayt
Contents
Editorial: Are You a Communist?
Essays of the Week
Who Is Shou Zi Chew, the TikTok CEO Trying to Reassure America?
U.S. Tech Should Support TikTok at This Moment
Banning TikTok Is Unconstitutional, Ludicrous, and a National Embarrassment
TikTok CEO says it wasn’t ‘spying’ when ByteDance employees surveilled journalists
Teens Social Media Habits and Experiences
Society’s Technical Debt and Software’s Gutenberg Moment
Amazon Faces Moment of Truth on Alexa as ChatGPT Steals Its Thunder
Video of the Week
The Age of AI and Our Human Future
News of the Week
OpenAI connects ChatGPT to the internet
OpenAI and Microsoft Are Partners, Until They Vie for the Same Customers
Apple reportedly plans to spend $1B a year to release Apple TV+ movies in theaters
Amazon layoffs will shut down camera review site DPReview.com after 25 years
JPMorgan to acquire Aumni, provider of venture capital analytics software
GitHub Copilot X: The AI-powered developer experience
The FTC Just Announced a Solution to an Annoying Problem
Startup of the Week
Character.AI Hits Unicorn Status With $150M Raise As AI Craze Continues
Tweet of the Week
Keith Teare (yes, me)
Editorial
Are You a Communist?
TikTok CEO Verbally Abused by Congress
When I woke up at about 7 am Pacific Time today, I turned on CNBC to see the CEO of TikTok testifying to Congress. It was an astounding experience, puzzling and worrying in equal proportions. Shou Zi Chew is a Singaporean. He speaks good English and is clear and thoughtful.
His questioners were like a pack of McCarthyite hounds to my eyes and ears. They grandstanded on their pet issues. They gave him no time to form answers. “Yes” or “No” was the most used phrase, and to questions that did not lend themselves to yes or no answers.
My heart sank as I listened to them try to accuse him of being an agent of the CCP, a threat to US national security., a proxy Communist, a teenage child abuser, and a spy.
All he could do was to sit, listen, vainly try to answer, and take it on the chin. He was really very gracious.
I have no idea if the questioners really believe their implied conspiracy theories. I hope the answer is no because they do not deserve their positions otherwise. But on the face of it, they believe that its Chinese origins make Tiktok prone to data theft from its US users and pass it to the Chinese Communist Party, that Tiktok is a security risk due to its connections with the Chinese government, that the platform has too much of a “grip” on American teenagers, causing social isolation among them, which has caused some to commit suicide due to the pressures of the platform.
While many American politicians would have you believe that TikTok needs to be banned from the United States, their primary reasons for suggesting such an action are invalid in all three of the main allegation categories. TikTok is not prone to data theft of US users, is not a security risk, and is not a breeding ground for suicide among teenagers. The claims are spurious and should be taken with a strong pinch of salt.
Data Theft
One of the most prominent accusations against TikTok is that its Chinese origins make it prone to data theft of its US users. However, this claim is not backed by substantial evidence. Like other social media platforms, TikTok collects user data to enhance the user experience and deliver targeted advertisements. The company has taken several steps to ensure data privacy, including storing user data in the United States and Singapore, with backups in the US.
Moreover, TikTok has implemented strict data access controls, ensuring that employees in China cannot access user data without approval from the US-based team. TikTok has also taken measures to be transparent about its data practices by publishing a Transparency Report detailing the number of government requests for user information and content removal. By doing so, TikTok demonstrates its commitment to data privacy, debunking the myth of data theft.
Security Risk Due to Connections with the Chinese Government
The second allegation against TikTok is that the platform poses a security risk due to its connections with the Chinese government. While TikTok’s parent company, ByteDance, is based in China, the company has taken numerous steps to distance itself from the Chinese government and safeguard user data.
In addition to storing data outside China, TikTok has hired a non-Chinese CEO, established a US-based security team, and engaged third-party experts to conduct audits of its data protection practices. The company has also established a Content Advisory Council, consisting of external experts who provide guidance on content moderation policies and practices. Now, with Project Texas, it is doing even more to firewall US Data.
It is important to note that no concrete evidence has emerged indicating that TikTok is a security risk or that the Chinese government has used the platform for espionage purposes. The claim that TikTok is a security risk due to its Chinese connections remains unsubstantiated. Shouting at the CEO really doesn’t change that.
TikTok’s Influence on American Teenagers
Lastly, TikTok has been accused of having too much of a “grip” on American teenagers, leading to social isolation and even suicide. However, evidence suggests that teen issues are predominantly caused by their real-world experiences, and social media platforms like TikTok are not the primary cause of these problems. In fact, social media can often serve as an avenue for teens to feel more engaged with each other and happier.
Research has shown that teenagers today face many pressures and challenges in their daily lives, such as academic stress, family issues, and societal expectations. These real-world experiences can contribute to feelings of anxiety, depression, and loneliness among teenagers. These platforms, including TikTok, can offer young people a sense of connection and belonging.
A study by the Pew Research Center found that 81% of teens aged 13-17 felt more connected to their friends through social media, and 68% received support during tough times through these platforms. TikTok, in particular, has emerged as a space where teenagers can express their creativity, share their interests, and connect with others who share their passions.
The short-form video format allows users to showcase their talents and engage in various challenges, fostering a sense of community among users. In addition, TikTok’s algorithm ensures that content is tailored to user’s preferences, making it more likely for them to find like-minded individuals and build connections.
Moreover, social media platforms like TikTok have been instrumental in facilitating conversations around mental health, helping to break the stigma associated with seeking help and support. Teenagers can access resources, connect with mental health advocates, and share their stories, creating a supportive online environment.
In conclusion, it is essential to consider the broader context when examining TikTok’s influence on American teenagers. While addressing the potential negative impacts of social media use is crucial, it is equally important to recognize the positive effects of platforms like TikTok on fostering connections and happiness among young people. By focusing on promoting responsible and healthy social media usage, we can ensure that TikTok remains a source of joy and connection for teenagers rather than an isolating force.
The facts are, in some ways, less interesting than the style of questioning. Congressmen and Congresswomen repeatedly used innuendo, accusations, badgering, shouting, and verbal abuse to intimidate Shou Zi Chew. It was not a spectacle of democracy in action, but that of a lynch mob, a show trial, if you will. To the ears of this immigrant, it was the worst of America displayed to a world audience. Shameful is not a strong enough word.
TikTok should be left alone to run its business. It is not accused of any law-breaking. America should not act like China, banning what it fears. And using fear of China to become like China is ironic.
I am with TikTok. And in this mode, I am not a friend of THIS Congress.
Essays of the Week
Who Is Shou Zi Chew, the TikTok CEO Trying to Reassure America?
Singapore army reservist and ex-Goldman banker says misconceptions about Chinese-owned app need to be clarified
By Stu Woo
WASHINGTON—TikTok Chief Executive Officer Shou Zi Chew has navigated both Western and Chinese business over the course of his fast rise to the corner office.
That cultural straddle has helped land him the top job at one of the world’s biggest tech companies. His background might now help him connect with a skeptical American Congress. Many members have threatened to ban TikTok, saying its Chinese ownership poses a national-security threat.
On Thursday, the 40-year-old Singaporean army reservist and former Goldman Sachs banker appeared before a House panel to argue against a ban and reassure Americans their data is safe and Beijing won’t be able to influence what viewers see on TikTok. Hours before the testimony, Beijing said it would oppose a forced sale of the app from its Chinese owners, something the Biden administration has demanded.
That amped up the hearing’s geopolitical dimensions, thrusting Mr. Chew into the middle of souring U.S.-China relations.
Born and raised in Singapore, then educated in London and at Harvard Business School, Mr. Chew spent some of his early career cutting deals for a venture-capital firm in Asia. He then quickly moved to the C-suite as the chief financial officer of a Chinese smartphone giant at age 32.
He runs TikTok from offices in Singapore, but travels often, including frequently to the U.S. He met his wife, a Taiwanese-American who grew up in Bethesda, Md., at Harvard.
Relatively unknown for the boss of one of the world’s most popular apps, Mr. Chew was sharply questioned by some of the company’s more skeptical critics on the House Energy and Commerce Committee. Aides to the Republicans who control the committee say there is essentially nothing Mr. Chew can say that would change their minds as they threaten a nationwide ban.
U.S. Tech Should Support TikTok at This Moment
By Martin Peers
March 22, 2023 5:00 PM PDT
·
If Steve Martin were to remake his 1984 film “Lonely Guy” this year, he could center it around Shou Zi Chew, CEO of TikTok. It’s a paradox that the executive leading one of the most popular apps in the U.S. is basically on his own as he fights the threat of a ban from both the Biden administration and congressional opponents. The few people who have publicly come out in support of TikTok include a couple of Democrats, the ACLU and some creators, such as a math teacher, as we reported today. Big tech firms? If they’ve taken a position publicly, it’s more likely to be against TikTok.
You might say fair’s fair—China doesn’t allow U.S. tech firms to operate freely, so why should the U.S. allow Chinese firms to do so? The flaw in that attitude, though, should be evident. The big U.S. tech companies typically generate around half their revenue from international markets. Do they really want to support anything that smacks of protectionism? Banning TikTok could set a precedent that other countries could use against U.S. tech firms. There’s also the fact that this anti-TikTok push reeks of xenophobia and double standards, as Julia Angwin eloquently said in a New York Times opinion piece on Monday. You’d think tech folks, who rely on foreign consumers and foreign talent, wouldn’t want to be associated with that kind of thinking.
Instead, short-term self-interest is prevailing. TikTok has proved a tough competitor to beat in the marketplace, both for its hold on consumer attention and increasingly for advertising dollars. Its rivals seem only too happy to see it banned. That came through clearly last September, when Snap CEO Evan Spiegel was asked at the Code conference whether TikTok should be banned. After observing that the government was reviewing the question of whether TikTok should be “in our country,” he noted the long-standing existence of restrictions on foreign ownership of U.S. broadcast media. More pointedly, though, he said the introduction of TikTok-like services by Snap, Meta Platforms and Google should resolve any government concerns about how a ban would affect TikTok fans.
In other words, Spiegel is saying it’s OK to ban TikTok because now American versions of the service are available. The backward, protectionist nature of the comment, coming from a tech executive who is selling to the world, was so startling that it’s hard to comprehend. We’ve seen similar strains of Sinophobia before: Not too long ago, tech companies were warning that China was going to dominate artificial intelligence. But the anti-TikTok campaign now underway, which carries a flavor of racist Yellow Peril commentary of the past, is something we’re likely to look back on with considerable regret.
How to Fix the TikTok Problem
March 20, 2023
By Julia Angwin
Ms. Angwin is a contributing Opinion writer and an investigative journalist.
Banning TikTok won’t keep us safe.
As you may have heard, the United States government is in the midst of a full-on panic about TikTok, the Chinese-owned video app best known for its ability to inspire teens to try out new dance moves.
Last week, the White House reportedly demanded that TikTok’s owners sell or face a ban. The ultimatum comes after months of anti-TikTok rhetoric: The head of the National Security Agency told Congress he is concerned about TikTok being used for foreign influence operations. A dozen senators introduced a White House-endorsed bill that would give the government broad powers to ban the app and other technology from China and five other unfriendly nations. Over two dozen states have passed legislation banning TikTok from government-owned devices. And the director of national intelligence said that parents should be worried about their kids using the app.
This week, Congress is hauling TikTok’s chief executive, Shou Zi Chew, in for a grilling about TikTok’s use of data about Americans, its impact on kids and its relationship with the Chinese Communist Party.
If only all the tech giants that prey on Americans’ data were getting the same scrutiny and enforced accountability. While Congress has been up in arms about TikTok, it has failed to pass even the most basic comprehensive privacy legislation to protect our data from being misused by all the tech companies that collect and mine it. Congress has also failed to follow the lead of Europe with its recent push to force platforms to be more accountable for the disinformation that they spread.
This whole episode is part of a larger red scare, in which the United States is taking an increasingly confrontational stance against China through economic sanctions imposed in the name of national security. U.S. tech executives and national security leaders have fed into this narrative, warning of an A.I. cold war in which China could surpass the United States in building artificial intelligence.
But when you dig into the national security allegations against TikTok, it is telling that most of the charges could just as easily be levied against the U.S. tech giants. And most of the tech companies’ exploitation of data has not been curbed by the government.
Let’s examine the concerns that have been raised about TikTok:….
Banning TikTok Is Unconstitutional, Ludicrous, and a National Embarrassment
It would also be an act of charity for Facebook, which the government has utterly failed to meaningfully regulate.
By Chloe Xiang, Jordan Pearson, Jason Koebler
American politicians are yet again obsessed with banning one of the most popular websites in the country. If Congress and the Biden administration follow through with banning TikTok, they will be violating the free speech of millions of creators, hamstringing millions of small (and large) businesses, and cutting Americans off from interacting with people around the world on a social network that is increasingly the cultural and meme engine of the internet. More cynically, they will be pissing off millions of younger Americans to satiate China hawks.
Banning TikTok would also be one of the biggest acts of charity toward a monopolistic American company in recent memory. It is not clear that Americans are any less safe on TikTok than they are on Meta’s platforms, which we know have been manipulated by foreign adversaries, make teens feel awful about themselves, allow their users to be targeted by anyone willing to pay, and have been proven to be used by white supremacists, genocidal regimes, etc.
TikTok CEO Shou Chew is testifying to Congress on Thursday in an effort to convince lawmakers that they should not ban the popular social media app in the U.S. “We do not believe that a ban that hurts American small businesses, damages the country’s economy, silences the voices of over 150 million Americans, and reduces competition in an increasingly concentrated market is the solution to a solvable problem,” Chew states in the prepared testimony that he will read in front of Congress.
This is a CEO who is trying to protect his own platform, just as Mark Zuckerberg and other social media giants did after their platform was used for Russian disinformation in the leadup to the 2016 election, and after their platforms were called monopolistic.
But banning Facebook or Twitter was never on the table, and it shouldn’t be on the table for TikTok, either. In fact, Congress has utterly failed to even meaningfully regulate these companies even after reams of studies have demonstrated their harm.
TikTok CEO says it wasn’t ‘spying’ when ByteDance employees surveilled journalists
Taylor Hatmaker
@tayhatmaker / 11:28 AM PDT•March 23, 2023
Image Credits: OLIVIER DOULIERY/AFP / Getty Images
At Thursday’s House hearing, TikTok CEO Shou Zi Chew faced a barrage of questions over concerns that data collected on U.S. users might be vulnerable to surveillance by China.
Chew’s response to one of those questions in particular likely dug the company’s hole even deeper when it comes to building trust with Washington.
Citing reporting from Forbes, later confirmed by the company itself, Florida Rep. Neal Dunn asked Chew if TikTok parent company ByteDance has spied on American citizens. Chew responded “I don’t think that spying is the right way to describe it.”
TEENS’ SOCIAL MEDIA HABITS AND EXPERIENCES
Teens and their experiences on social media
BY MONICA ANDERSON AND JINGJING JIANG
PEW RESEARCH CENTER
NOVEMBER 28, 2018
Social media has given teens the ability to instantly connect with others and share their lives through photos, videos and status updates. Teens themselves describe these platforms as a key tool for connecting and maintaining relationships, being creative, and learning more about the world. But they also must contend with more negative aspects of social media use, such as drama and bullying or feeling pressure to present themselves in a certain way.
Teens post about a range of topics on social media, with posts about their accomplishments or family playing an especially prominent role
When asked what topics they post about on social media, roughly half of teens say they post about their accomplishments on social media, while 44% say they post about their family. Around one-third (34%) say they share things related to their emotions and feelings on these sites, while 22% report posting about their dating life. Relatively few teens – around one-in-ten – say they share things related to their personal problems or their religious or political beliefs on social media.
There are some age and gender differences in the topics teens share on social media. Older teens are more likely than their younger counterparts to post about their romantic relationships: 26% of teens ages 15 to 17 say they post about their dating life on social media, compared with 16% of 13- to 14-year-olds.
Meanwhile, girls are more likely than boys to say they post about their family (53% vs. 36%), their emotions and feelings (40% vs. 29%) or their religious beliefs (14% vs. 7%). And older girls are especially likely to post about a variety of subjects – including their dating lives, their family, their emotions and their religious or political beliefs, compared with older boys or younger teens.
Selfies may be popular on social media, but around half of teens say they rarely or never post these images
Although the proliferation of smartphones has given teens the ability to constantly share different aspects of their lives, this survey finds that many teens regularly forego posting selfies, videos or other updates of their lives to social media.
Society’s Technical Debt and Software’s Gutenberg Moment
Mar 21
Irregular Ideas with Paul Kedrosky & Eric Norlin of SKV
Abstract
There is immense hyperbole about recent developments in artificial intelligence, especially Large Language Models like ChatGPT. And there is also deserved concern about such technologies’ material impact on jobs. But observers are missing two very important things:
Every wave of technological innovation has been unleashed by something costly becoming cheap enough to waste.
Software production has been too complex and expensive for too long, which has caused us to underproduce software for decades, resulting in immense, society-wide technical debt.
This technical debt is about to contract in a dramatic, economy-wide fashion as the cost and complexity of software production collapses, releasing a wave of innovation.
*****************
Software is misunderstood. It can feel like a discrete thing, something with which we interact. But, really, it is the intrusion into our world of something very alien. It is the strange interaction of electricity, semiconductors, and instructions, all of which somehow magically control objects that range from screens to robots to phones, to medical devices, laptops, and a bewildering multitude of other things. It is almost infinitely malleable, able to slide and twist and contort itself such that, in its pliability, it pries open doorways as yet unseen.
This sense of the alien-other nature of software has bubbled up to the surface of public consciousness recently, as truly conversational artificial intelligence—based on Large Language Models (LLMs), such as ChatGPT—have gone from the stuff of science fiction to something that people can play with as easily as they can search the internet.
Why Hasn’t Software Eaten the World?
This can feel reminiscent of Marc Andreessen’s line from the last tech cycle, that “software is eating the world.” And while Marc’s line is a useful way of thinking about software’s appetites, it is worth wondering why software is taking so damn long to finish eating. And, further, if it were to really get to feasting, what would be the catalyst or catalysts, and what would that software-eaten world look like?
To answer these questions, we must look back before we look forward. In doing so, we will consider a range of factors, including complexity, factor costs, and economic models of software supply and demand.
Let’s start by considering an economic model of software supply and demand. Software has a cost, and it has markets of buyers and sellers. Some of those markets are internal to organizations. But the majority of those markets are external, where people buy software in the form of apps, or cloud services, or games, or even embedded in other objects that range from Ring doorbells to endoscopic cameras for cancer detection. All of these things are software in a few of its myriad forms.
With these characteristics in mind, you can think of software in a basic price/quantity graph from introductory economics. There is a price and a quantity demanded at that price, and there is a price and quantity at which those two things are in rough equilibrium, as the following figure shows. Of course, the equilibrium point can shift about for many reasons, causing the P/Q intersection to be at higher or lower levels of aggregate demand. If the price is too high we underproduce software (leaving technical debt), and if too low, well … let’s come back to that.
This leads us to a basic question that sometimes gets asked in economics classes: How do we know this combinate of price and quantity is the optimal one? The orthodox answer is that if the the supply and demand lines are correct, then it must be the right one, given the downward-sloping demand (people want less of most things when the price goes up), and the upward-sloping supply (manufacturers supply more of a thing when the price goes up). When we go past where those two lines intersect, every increment in price might make manufacturers want to produce more, but leave buyers wanting less, so you end up with excessive inventories or prices crashes, or both. And, in reverse, when the price falls, manufacturers want to produce less, even if buyers want more, which leads to scarcity, and, in turn, often drives prices higher again until supply and demand equilibrate.
How Technology Confounds Economics
This is all straightforward undergraduate economics. But technology has a habit of confounding economics. When it comes to technology, how do we know those supply and demand lines are right? The answer is that we don’t. And that’s where interesting things start happening.
Sometimes, for example, an increased supply of something leads to more demand, shifting the curves around. This has happened many times in technology, as various core components of technology tumbled down curves of decreasing cost for increasing power (or storage, or bandwidth, etc.). In CPUs, this has long been called Moore’s Law, where CPUs become more powerful by some increment every 18 months or so. While these laws are more like heuristics than F=ma laws of physics, they do help as a guide toward how the future might be different from the past.
We have seen this over and over in technology, as various pieces of technology collapse in price, while they grow rapidly in power. It has become commonplace, but it really isn’t. The rest of the economy doesn’t work this way, nor have historical economies. Things don’t just tumble down walls of improved price while vastly improving performance. While many markets have economies of scale, there hasn’t been anything in economic history like the collapse in, say, CPU costs, while the performance increased by a factor of a million or more.
To make this more palpable, consider that if cars had improved at the pace computers have, a modern car would:
Have more than 600 million horsepower
Go from 0-60 in less than a hundredth of a second
Get around a million miles per gallon
Cost less than $5,000
And they don’t. Sure, Tesla Plaid is a speedy car, it is nowhere near the above specs—no car ever will be. This sort of performance inflection is not our future, but it fairly characterizes and even understates what has happened in technology over the last 40 years. And yet, most people don’t even notice anymore. It is just commonplace, to the point that our not noticing is staggering.
The Collapse Dynamic in Technology
You can see these dynamics in the following graph. Note the logarithmic scale on the Y-axis, which prevents the price/performance lines from being vertical—that is how quickly and sharply the price per unit performance of all these factors have fallen. It is unprecedented in economic history.
And each of these collapses has had broader consequences. The collapse of CPU prices led us directly from mainframes to the personal computer era; the collapse of storage prices (of all kinds) led inevitably to more personal computers with useful local storage, which helped spark databases and spreadsheets, then led to web services, and then to cloud services. And, most recently, the collapse of network transit costs (as bandwidth exploded) led directly to the modern Internet, streaming video, and mobile apps.
In a kind of reversal of the old Paul Simon song (“The Boy in the Bubble”), every technology generation throws a hero down the pop (or, price per unit performance) charts. Each collapse, with its accompanying performance increases, sparks huge winners and massive change, from Intel, to Apple, to Akamai, to Google & Meta, to the current AI boomlet. Each beneficiary of a collapse requires one or more core technologies’ price to drop and performance to soar. This, in turn, opens up new opportunities to “waste” them in service of things that previously seemed impossible, prohibitively expensive, or both.
AI as the Next Collapse Dynamic in Technology
All of that brings us to today. Suddenly AI has become cheap, to the point where people are “wasting” it via “do my essay” prompts to chatbots, getting help with microservice code, and so on. You could argue that the price/performance of intelligence itself is now tumbling down a curve, much like as has happened with prior generations of technology.
You could argue that, but it’s too narrow and orthodox a view, or at the very least, incomplete and premature. Set aside the ethics and alignment issues of artificial general intelligence (AGI). It is likely still years away, at best, even if it feels closer than it has in decades. In that light, it’s worth reminding oneself that waves of AI enthusiasm have hit the beach of awareness once every decade or two, only to recede again as the hyperbole outpaces what can actually be done. We saw this in the 1950s with Minsky’s (failed) work, again in the 1970s with Japan’s (failed) Fifth Generation Project, and again in the 2000s with IBM’s (failed) Watson. If you squint really hard, you might see a pattern.
The Age of AI has begun
Artificial intelligence is as revolutionary as mobile phones and the Internet.
By Bill Gates
March 21, 2023
In my lifetime, I’ve seen two demonstrations of technology that struck me as revolutionary.
The first time was in 1980, when I was introduced to a graphical user interface—the forerunner of every modern operating system, including Windows. I sat with the person who had shown me the demo, a brilliant programmer named Charles Simonyi, and we immediately started brainstorming about all the things we could do with such a user-friendly approach to computing. Charles eventually joined Microsoft, Windows became the backbone of Microsoft, and the thinking we did after that demo helped set the company’s agenda for the next 15 years.
The second big surprise came just last year. I’d been meeting with the team from OpenAI since 2016 and was impressed by their steady progress. In mid-2022, I was so excited about their work that I gave them a challenge: train an artificial intelligence to pass an Advanced Placement biology exam. Make it capable of answering questions that it hasn’t been specifically trained for. (I picked AP Bio because the test is more than a simple regurgitation of scientific facts—it asks you to think critically about biology.) If you can do that, I said, then you’ll have made a true breakthrough.
I thought the challenge would keep them busy for two or three years. They finished it in just a few months.
In September, when I met with them again, I watched in awe as they asked GPT, their AI model, 60 multiple-choice questions from the AP Bio exam—and it got 59 of them right. Then it wrote outstanding answers to six open-ended questions from the exam. We had an outside expert score the test, and GPT got a 5—the highest possible score, and the equivalent to getting an A or A+ in a college-level biology course.
Once it had aced the test, we asked it a non-scientific question: “What do you say to a father with a sick child?” It wrote a thoughtful answer that was probably better than most of us in the room would have given. The whole experience was stunning.
I knew I had just seen the most important advance in technology since the graphical user interface.
This inspired me to think about all the things that AI can achieve in the next five to 10 years.
The development of AI is as fundamental as the creation of the microprocessor, the personal computer, the Internet, and the mobile phone. It will change the way people work, learn, travel, get health care, and communicate with each other. Entire industries will reorient around it. Businesses will distinguish themselves by how well
Amazon Faces Moment of Truth on Alexa as ChatGPT Steals Its Thunder
Five years ago, Amazon’s Alexa voice assistant had it all: high-profile partnerships, booming device sales and plenty of industry and pop culture buzz. But ChatGPT has grabbed the spotlight in AI at a delicate moment when critics say Alexa has stagnated.
Art by Mike Sullivan
By Theo Wayt
March 22, 2023 6:05 AM PDT
At the 2018 Consumer Electronics Show, Amazon announced one of its biggest partnerships yet to help make its Alexa voice assistant ubiquitous: a deal with Toyota to integrate Alexa into the auto giant’s cars. “Our vision for Alexa is that she should be everywhere a customer might need her—at home, in the office, on phones—and in cars,” an Amazon executive gushed in a press release about the agreement.
Toyota doesn’t seem to need Alexa anymore. The automaker has dropped support for an app that allowed users to operate Alexa in their cars via smartphones in 2023 editions of several of its most popular models, including the RAV4, Prius and Corolla. And according to a person close to the automaker, Toyota plans to phase out Alexa integration from the rest of its lineup in the coming years. The automaker is now focused on improving an in-house voice assistant it launched last year and is considering integrating ChatGPT, the chatbot created by OpenAI, into it, the person said.
Toyota’s breakup with Alexa, details of which haven’t been reported previously, is a potent symbol of the challenges facing Amazon’s once-mighty voice assistant. Despite selling hundreds of millions of digital devices with the assistant built into them, Alexa has fallen short of Amazon’s goal of creating the next big platform in tech. Even onetime believers in the technology say innovation has stalled on Alexa, a sentiment that has grown since ChatGPT set the tech world ablaze when it came out in November.
THE TAKEAWAY
• Toyota has dropped support for Alexa from popular vehicles
• Amazon is exploring partnerships with AI startups
• The talks have involved bringing ChatGPT-like functions to Alexa, product search
ChatGPT has been hailed as a breakthrough for artificial intelligence applications, one that can converse persuasively with users (in text form, for now), write software code and perform other tasks. What’s more, the startup responsible for it, OpenAI, has done so with roughly 375 employees. That’s far fewer than the 10,000 employees devoted to Alexa-related projects, a figure that explains the ongoing companywide layoffs at Amazon, which hit the Alexa team early and especially hard.
Increasing the stakes for Amazon is the fact that its biggest rival in cloud computing, Microsoft, has poured billions of dollars in investment into OpenAI and has begun weaving its technology—based on a type of software known as a large-language model—into the Bing search engine and the Microsoft 365 suite of productivity applications.
“It surprised me that OpenAI and Microsoft beat them to the jump,” said Tigger Kindel, a former Amazon executive who spent five years working on Alexa. “They’ve been kind of refining the same old thing and now it’s 2023.”
Video of the Week
News of the Week
OpenAI connects ChatGPT to the internet
Kyle Wiggers
@kyle_l_wiggers / 10:35 AM PDT•March 23, 2023
Image Credits: OpenAI
OpenAI’s viral AI-powered chatbot, ChatGPT, can now browse the internet — in certain cases.
OpenAI today launched plugins for ChatGPT, which extend the bot’s functionality by granting it access to third-party knowledge sources and databases, including the web. Available in alpha to ChatGPT users and developers on the waitlist, OpenAI says that it’ll initially prioritize a small number of developers and subscribers to its premium ChatGPT Plus plan before rolling out larger-scale and API access.
Easily the most intriguing plugin is OpenAI’s first-party web-browsing plugin, which allows ChatGPT to draw data from around the web to answer the various questions posed to it. (Previously, ChatGPT’s knowledge was limited to dates, events and people prior to around September 2021.) The plugin retrieves content from the web using the Bing search API and shows any websites it visited in crafting an answer, citing its sources in ChatGPT’s responses.
OpenAI and Microsoft Are Partners, Until They Vie for the Same Customers
By Aaron Holmes
March 21, 2023 6:00 AM PDT ·
OpenAI’s ChatGPT has enraptured the business world since its November release and OpenAI is signing up customers eager to pay to use its artificial intelligence models in their own products. But the Microsoft-backed startup faces a surprising rival: Microsoft itself.
As part of its multibillion-dollar investment in OpenAI, Microsoft has the rights to sell the startup’s software through its Azure cloud business, even as OpenAI licenses its own software directly to customers. Microsoft also gets a share of OpenAI’s profits. The offerings cost the same, a fraction of a cent per query. Meanwhile, all of OpenAI’s technology runs on Microsoft’s Azure cloud infrastructure rent free.
The dual offerings mean the companies are at times pitching the same customers on nearly identical products, putting salespeople at Microsoft in the uneasy position of trying to lure customers away from OpenAI while touting its technology.
THE TAKEAWAY
• Microsoft and OpenAI are both marketing OpenAI licenses, sometimes to the same customers
• Microsoft has told its salespeople to tout Azure’s security features as an advantage over OpenAI
• OpenAI is ramping up its enterprise deals, including with Microsoft competitors
While the profit-sharing agreement means sales of either offering theoretically benefit both companies, OpenAI pursues direct relationships with big customers, such as Microsoft rival Salesforce, which has licensed ChatGPT for a new suite of customer service software. It’s not clear whether the partnership between OpenAI and Microsoft dictates the price each company sets for the models. Microsoft gets 75% of OpenAI’s profits until its investment is paid back and 49% of subsequent profits up to a certain cap, The Information previously reported. It’s also not clear how much profit Microsoft returns to OpenAI for the models it sells through Azure OpenAI Service. Representatives for Microsoft and OpenAI did not immediately provide comment.
OpenAI vs. Azure OpenAI
Microsoft is eager to use OpenAI’s technology to lure cloud customers away from competitors like Amazon and Google, as well as prod existing cloud customers to spend more on Azure. Through the OpenAI partnership, Microsoft is the only major cloud provider to offer access to large-language models that customers can use in products such as customer service chatbots, search engines or fraud detection algorithms. That gives it an important advantage over competitors. (Google announced last week that it would soon make its own large-language model available through a similar service).
Convincing customers to pay for bigger bundles of cloud services is a key pillar of Microsoft’s strategy to boost Azure growth amid a broader slowdown in cloud spending. Microsoft touts Azure’s security and compliance features, as well as its reputation for dealing with big cloud customers, according to one current Microsoft salesperson and one former employee.
Apple reportedly plans to spend $1B a year to release Apple TV+ movies in theaters
Lauren Forristal
@laurenforristal / 8:18 AM PDT•March 23, 2023
Image Credits: Apple
Apple may be partnering with studios to put more titles in theaters, a plan that could cost $1 billion a year. Typically, Apple’s original movies have either received small theatrical runs or been exclusive to Apple TV+. So, the potential investment would be a notable increase and signals a major shift in the streamer’s strategy.
According to Bloomberg, which reported the news, anonymous sources say that the Apple plans aren’t finalized yet. However, the company wants to put films in thousands of theaters for at least one month.
Apple declined to comment to TechCrunch. The company also declined to comment to Bloomberg.
Some possible releases, Bloomberg wrote, include Martin Scorsese’s “Killers of the Flower Moon,” starring Leonardo DiCaprio; Ridley Scott’s “Napoleon” and the spy thriller “Argylle.”
However, the Scorsese film is somewhat separate, reported 9to5Mac, as Apple apparently paid about $200 million on the production rights from Paramount a couple of years ago. Paramount will distribute “Killers of the Flower Moon” to cinemas worldwide for a few months before the title premieres exclusively on Apple TV+.
Amazon layoffs will shut down camera review site DPReview.com after 25 years [Updated]
Updates stop on April 10, site will be available for “a limited period” after.
ANDREW CUNNINGHAM – 3/21/2023, 11:47 AM
Amazon has plans to lay off at least 27,000 workers this year, including 9,000 that were announced in an internal email Monday morning. One unexpected casualty: Digital Photography Review, also known as DPReview, is losing its entire editorial staff, and the site will stop publishing on April 10.
The announcement post, written by DPReview General Manager Scott Everett, says that new pieces will continue to be posted through April 10, and “the site will be locked” afterward. It’s unclear what will happen to the site’s content afterward—the post promises only that the site’s articles “will be available in read-only mode for a limited period afterwards.” Any photos and text that readers have uploaded to their accounts can be requested and downloaded until April 6, “after which we will not be able to complete the request.”
Former site editor Gannon Burgett said on Twitter that the decision to lay off the staff was announced in January and that “Amazon hasn’t yet come up with an archival plan” for the site. Cameras, even digital ones, tend to have a pretty long shelf life, and there’s an active used market for lenses and camera bodies—if DPReview.com goes offline entirely, that would be a huge blow to anyone trying to research older products.
Founded in November 1998, DPReview is one of the few active review sites as old as Ars Technica. Amazon purchased it in 2007, and the site’s team has been located in Amazon’s hometown of Seattle since 2010.
JPMorgan to acquire Aumni, provider of venture capital analytics software
Mar. 22, 2023 10:29 AM ETJPMorgan Chase & Co. (JPM)By: Liz Kiesche, SA News Editor
JPMorgan Chase (NYSE:JPM) agreed on Wednesday to acquire Aumni, a provider of investment analytics software for the venture capital industry, part of its drive to bolster its technology capabilities.
Financial terms of the deal weren’t disclosed, though CNBC reported that J.P. Morgan (JPM) will pay about what the firm was valued for at its last fundraising in 2021. Pitchbook valued the five-year-old Aumni at $232M after that round, CNBC said.
This acquisition specifically beefs up the Wall Street bank’s private markets platform and offerings for venture capital clients. The deal comes after JPMorgan’s recent launch of Capital Connect and its acquisition of Global Shares.
Aumni’s data analytics engine structures, tracks, and analyzes legal and economic terms underpinning growth-stage private market transactions. With more than 300 clients, the company has evaluated more than $600B in invested capital across more than 17,000 private companies, J.P. Morgan said.
About a year ago, JPMorgan (JPM) agreed to acquire Global Shares, a cloud-based provider of share plan management software.
While JPMorgan (JPM) has been an active acquirer of fintechs to give it a competitive edge, the Wall Street giant claimed that it was duped in its 2021 acquisition of college financial site Frank.
GitHub Copilot X: The AI-powered developer experience
GitHub Copilot is evolving to bring chat and voice interfaces, support pull requests, answer questions on docs, and adopt OpenAI’s GPT-4 for a more personalized developer experience.
Author: Thomas Dohmke
March 22, 2023
At GitHub, our mission has always been to innovate ahead of the curve and give developers everything they need to be happier and more productive in a world powered by software. When we began experimenting with large language models several years ago, it quickly became clear that generative AI represents the future of software development. We partnered with OpenAI to create GitHub Copilot, the world’s first at-scale generative AI development tool made with OpenAI’s Codex model, a descendent of GPT-3.
GitHub Copilot started a new age of software development as an AI pair programmer that keeps developers in the flow by auto-completing comments and code. And less than two years since its launch, GitHub Copilot is already writing 46% of code and helps developers code up to 55% faster.
But AI-powered auto-completion is just the starting point. Our R&D team at GitHub Next has been working to move past the editor and evolve GitHub Copilot into a readily accessible AI assistant throughout the entire development lifecycle. This is GitHub Copilot X—our vision for the future of AI-powered software development. We are not only adopting OpenAI’s new GPT-4 model, but are introducing chat and voice for Copilot, and bringing Copilot to pull requests, the command line, and docs to answer questions on your projects.
With AI available at every step, we can fundamentally redefine developer productivity. We are reducing boilerplate and manual tasks and making complex work easier across the developer lifecycle. By doing so, we’re enabling every developer to focus all their creativity on the big picture: building the innovation of tomorrow and accelerating human progress, today.
Let’s jump in.
Want to see what’s new? Discover GitHub Copilot X—our vision for the future of AI-powered software development. Learn more >
A new AI-powered developer experience
A ChatGPT-like experience in your editor with GitHub Copilot Chat: We are bringing a chat interface to the editor that’s focused on developer scenarios and natively integrates with VS Code and Visual Studio. This does far more than suggest code. GitHub Copilot Chat is not just a chat window. It recognizes what code a developer has typed, what error messages are shown, and it’s deeply embedded into the IDE. A developer can get in-depth analysis and explanations of what code blocks are intended to do, generate unit tests, and even get proposed fixes to bugs.
GitHub Copilot Chat builds upon the work that OpenAI and Microsoft have done with ChatGPT and the new Bing. It will also join our voice-to-code AI technology extension we previously demoed, which we’re now calling GitHub Copilot Voice, where developers can verbally give natural language prompts.
ADOBE LAUNCHES FIREFLY AI – NO UNLICENSED PHOTOS FOR TRAINING
Mar 21, 2023 by Dunja Djudjic 2 Comments
Adobe has launched Adobe Firefly, yet another text-to-image tool that lets you turn any set of words you can think of into a picture, video, 3D model, and more. After some criticism, Adobe pointed out that it didn’t use any of your stored photos to train its algorithms. It relied only on the work licensed on Adobe Stock.
WHAT IS ADOBE FIREFLY?
Let’s start by presenting you with Adobe Firefly and what it’s capable of. Judging from the demo, it can do pretty much anything you can think of. It works like any other text-to-image generator: you type in your prompts and wait for the image to come to life on your screen. From there on, you can use a brush to paint away and modify any bit of the picture. You can also expand its borders, similar to DALL-E. Other than words alone, you can also use 3D models and turn them into pictures; or imagine anything and then turn it into a video.
And in addition to pictures, you can use textual prompts to create patterns, vectors, brushes, and templates. You can also take a sketch and turn it into a full-color image. It reminded me of Adobe Scribbler introduced back in 2017, and I assume that it builds upon it. All in all, Adobe Firefly seems like a pretty versatile and useful text-to-image tool.
ADOBE’S ALGORITHM TRAINING
“We have never, ever used anything in our storage to train a generative AI model, not once,” Adobe’s Chief Product Officer Scott Belsky said in an interview with Bloomberg earlier this year. He said that the company’s decade-old Terms of Service were misinterpreted, as they were in place “to allow products to be analyzed to improve features, not for image generation.” Belsky added that Adobe was rolling out a new iteration of this policy that is more specific. “If we ever allow people to opt in for generative AI specifically, we need to call it out and explain how we’re using it,” he said.
The FTC Just Announced a Solution to an Annoying Problem
It could become harder for companies to trick you into keeping a subscription you don’t want.
If you’ve ever checked a bank statement and realized you’ve been paying for a monthly subscription for four months that you meant to cancel after a free trial, good news — the FTC is looking out for you.
On Thursday, the Federal Trade Commission proposed making a rule change that would make it easier to cancel free trials and subscriptions customers no longer want.
“The idea here is pretty simple. Companies should not be able to manipulate consumers into paying for subscriptions that they don’t want,” FTC Chair Lina Khan said in a conference call Wednesday, according to CNN.
Under the new proposal companies would be barred from trying to retain customers through nefarious or burdensome tactics that intended to keep them from quitting a subscription.
It would also require companies to first ask consumers who are trying to cancel a free subscription whether they would be open to retention offers before trying to entice them.
Products subject to the new rule would include gym memberships, digital streaming and e-commerce, cable TV service, traditional print media and others.
The regulatory body voted 3-1 to introduce the proposal and said that the public will have an opportunity to comment on the proposal.
Startup of the Week
Character.AI Hits Unicorn Status With $150M Raise As AI Craze Continues
March 23, 2023
At this point, you can almost set your clock by it — every week there is a huge raise by an AI startup.
This week — like the last — the big round created a new unicorn as Character.AI closed a $150 million Series A at a $1 billion valuation led by Andreessen Horowitz. The round had been reported earlier this month.
The round also included participation from previous investors, including Nat Friedman — former GitHub CEO — Elad Gil, SV Angel and A Capital.
The Palo Alto, California-based AI startup allows people to create their own personalized AI chatbot using language models and deep-learning algorithms. The AI-created companions can help users draft emails, serve as a study buddy, brainstorm ideas or a variety of other activities.
Investors just can’t get enough
Of course, Character.AI is far from the only AI startup this year to raise cash. Despite strong headwinds — like a drop in venture financing and the collapse of Silicon Valley Bank — VCs and large strategics including Microsoft, Google and Salesforce – all continue to show unabated interest in generative AI startups.