Tech Brew Ride Home - Fri. 10/06 – Microsoft Looks To Close

Episode Date: October 6, 2023

Looks like Microsoft is inches away from getting the Activision acquisition over the line. Hardware ambitions are one thing, but OpenAI might also design its own silicon. Microsoft’s new version of Teams is finally not the most resource hungry piece of software on your computer. And, of course, the Weekend Longreads Suggestions. Links: Microsoft eyes closing its giant Activision Blizzard deal next week (The Verge) Exclusive: ChatGPT-owner OpenAI is exploring making its own AI chips (Reuters) Microsoft’s faster and redesigned Teams app now available for Windows and Mac (The Verge) Weekend Longreads Suggestions: How Will A.I. Learn Next? (The New Yorker) How Bandcamp makes more money than Spotify (Fast Company) Long gone, DEC is still powering the world of computing (ArsTechnica) The Genius Behind Hollywood’s Most Indelible Sets (NYTimes Magazine) Learn more about your ad choices. Visit megaphone.fm/adchoices

On April 4th, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. Welcome to the Tech meme right home for Friday, October 6, 23. I'm Brian McCullough today. Looks like Microsoft is inches away from getting the Activision acquisition over the line. Hardware ambitions are one thing, but OpenAI might also design its own silicon.

Microsoft's new version of Teams is finally not the most resource-hungry piece of software on your computer. And of course, the weekend long-read suggestions. Here's what you miss today in the world of tech. Looks like we might be able to close the loop on this one at long last. A source is reporting that Microsoft plans to close its $68.7 billion Activision acquisition a week from today, October 13th, with a final decision from the UK's CMA allowing the deal expected to come down next week. Quoting the verge.

That date will still depend on the UK's competition and markets authority, though, a regulator that blocked Microsoft's deal earlier this year. Microsoft recently restructured the deal to transfer cloud gaming rights for current and new Activision Blizzard games to Ubisoft and and the Xbox maker secured preliminary approval from the CMA late last month as a result. The CMA has a deadline that expires today on gathering opinions over whether it should grant consent to Microsoft to proceed with the merger. A final decision from the CMA is expected next week, and barring any surprise, last-minute changes should allow Microsoft to close the deal.

Microsoft and Activision extended their deal deadline to October 18th recently, but if Microsoft is able to close the deal next week, it will bring to a close a 20-month process of regulatory approvals and battles across Europe and the U.S. a little earlier than expected. Here in the U.S., the FTC is still appealing an outcome of a hearing with the Ninth Circuit Court of Appeals, and a decision on that is due in early December. The FTC is also planning to resume its own administrative case against Microsoft's proposed Activision Blizzard acquisition. The administrative case will commence 21 days after the Ninth Circuit rules on the FTC's appeal, with the hearing held virtually. The FTC could attempt to undo Microsoft's deal, assuming it closes on time, but it would

face an unprecedented uphill battle, end quote. I find the rumored ambitions of OpenAI to be hella interesting. We've heard of them thinking they need their own hardware platform, but what if they also started doing their own chips? Quoting Reuters. Open AI, the company behind ChatGBT is exploring making its own artificial intelligence chips and has gone as far as evaluating a potential acquisition target, according to people familiar with the company's plans. The company has not yet decided to move ahead, according to recent internal discussions described to Reuters. However, since at least last year, it discussed various options to solve the shortage of expensive AI chips that OpenAI relies on, according to people familiar with the matter. These options have included

building its own AI chip, working more closely with other chipmakers, including Nvidia, and also diversifying its suppliers beyond Nvidia. The effort to get more chips is tied to two major concerns CEO Sam Altman has identified, a shortage of the advanced processors that power OpenAI software and the, quote, eye-watering costs associated with running the hardware necessary to power its efforts and products. Since 2020, OpenAI has developed its generative artificial intelligence technologies on a massive supercomputer constructed by Microsoft, one of its largest backers that uses 10,000 of Nvidia's graphics processing units or GPUs. Running chat GPT is very expensive for the company. Each query costs roughly four cents, according to an analysis from Bernstein analyst Stacey Razgun.

If chat GPT queries grow to a tenth the scale of Google search, it would require roughly $48.1 billion worth of GPUs initially at about $16 billion worth of chips a year to keep operational. It's not clear whether OpenAI will move ahead with a plan to build a custom chip. Doing so would be a major strategic initiative and a heavy investment that could amount to hundreds of millions of dollars a year in cost, according to industry veterans. Even if OpenAI committed resources to the task, it would not guarantee success. An acquisition of a chip company could speed the process of building OpenAI's own chip, as it did for Amazon and its acquisition of Anna Perna Labs in 2015. OpenAI had considered the path to the point where it performed due

diligence on a potential acquisition target, according to one of the people familiar with its plans. The identity of the company OpenAI examined purchasing could not be learned. Even if Open AI goes ahead with plans for a custom chip, including an acquisition, the effort is likely to take several years, leaving the company dependent on commercial providers like Nvidia and advanced micro devices in the meantime, end quote. You might have noticed this, but Microsoft has rolled out its new version of Microsoft Teams, which has been in preview since March, rebuilt from the ground up to be less resource hungry on Windows and MacOS. quoting the verge. The new version of Teams is up to two times faster while using 50% less memory according to Microsoft. That's good news to anyone who uses Teams already, which has been particularly resource-hungry on older laptops and PCs. Installing the new Microsoft Teams app should be three times faster and launching or joining meetings two times faster with the app now taking up 70% less disk space.

All of these performance improvements are thanks to Microsoft ditching the electron foundations of Teams. and moving toward Microsoft's Edge WebView2 technology. Microsoft has also moved to the React JavaScript library and focused on improving the Microsoft Teams design with the fluent design language system for several UI improvements. The new Teams has been in preview for months, but during that time, it was missing some features that Microsoft has now added to the final version.

We have made notable progress since the launch of new teams in public preview, said Microsoft product lead, anupam Patnake. Quote, new teams now has, full feature parity for almost all features, including custom line of business apps, third-party apps, breakout rooms, seven-by-seven video, call cues, PSTN calling, contextual search in chats and channels, cross-post a channel conversation, and more, end quote. New Microsoft Teams features will now be delivered exclusively on this new Teams client, so businesses will need to upgrade. There's no migration required, though, so upgrading

should be as simple as an update. Microsoft says classic Teams users will be automatically upgraded to new teams in the coming months. The new Teams client also wasn't available initially for Mac users earlier this year in preview, but with the final release for MacOS, there are also improvements for Apple's devices. We're also seeing significant performance improvements on Mac, including the ability to switch between chats and channels faster, and access relevant information quickly and efficiently, with a faster scrolling experience, says Microsoft. Teams works natively on all Mac devices, including those with Apple Silicon, giving Mac users an improved app experience. This new Microsoft Teams client is also the foundation for the company's AI-powered

copilot push in Teams. You'll be able to use co-pilot in Teams to summarize meetings, read action items, and avoid long threads of conversations to get to the key points, end quote. Time for the weekend long read suggestions. First up, the New Yorker has a deep dive into a question and we keep coming back to. If AI has been trained on the web, but the web is going to be run over soon by tons of AI content, where will AI go to teach itself in the future? Quote, a major leap in AI may come when LLMs start seeming curious or bored. Curiosity and boredom sound like they belong to an organic mind, but here's how they might be created inside an AI. As a rule, chatbots today have a propensity to confidently make stuff up,

or, as some researchers say, hallucinate. At the root of these hallucinations is an inability to introspect. The AI doesn't know what it does and doesn't know. As researchers begin to solve the problem of getting their models to express confidence and cite their sources, they will not just be making chatbots more credible, they will also be equipping them with a rudimentary kind of self-knowledge. An AI will be able to observe from reams of its own chat transcripts that it is prone to hallucination in a particular area. It will only be natural. to let that tendency guide its ingestion of further training data. The model will direct itself towards sources that touch on topics it knows the least about, curiosity in its most basic form.

If it can't find the right kind of training data, a chatbot might solicit it. I imagine a conversation with some future version of chat GPT in which, after a period of inactivity, it starts asking me questions. Perhaps, having observed my own questions and follow-ups, it will have developed an idea of what I know about. You're a programmer and a writer, aren't you? It might say to me, sure, I'll respond. I thought so. I'm trying to get better at technical writing. I wonder if you could help me decide which of the following sentences is best. Such an AI might ask my sister who works at a construction company about what's going on in the local lumber market. It could ask my doctor friend who does research on cancer, whether he could clear up something in a recent nature paper. Such a system would be like

Stack Overflow, Wikipedia, and Reddit combined. Except that instead of knowledge getting deposited into the public square, it would accumulate privately in the mind of an ever-growing genius. Observing the web collapsed this way into a single gigantic chatbot would be a little like watching a galaxy spiral into a black hole, end quote. Then, Fast Company takes a look at an interesting question, why is Bancamp profitable, but Spotify, after all these years, still isn't. Quote, Bancamp bakes heterogeneity into its platform in a way that Spotify's one-size-fits-all-service erases. So while Spotify has created a technically complex platform for simple transactions,

Ban Camp has created a technically simple platform for complex transactions. The result is that Ban Camp is able to reap the benefits of what's known as fat-tailed distributions, in which a minority of individuals comprise the majority of sales. The same logic followed by venture capital firms whose bets on one or two unicorns make up for all the other portfolio companies that go bust. As I found in a recent analysis, about 20% a band camp customers account for 80% of the site's total revenue. And within this 20%, the sale of physical goods becomes increasingly important. More than half the objects that the very top spenders buy on the site are physical. And among those objects, vinyl records are more popular

than all others, including CDs, tapes, and miscellaneous merch combined. On average, artists who sell the most physical stuff make the most money on the platform, and among the different kinds of physical stuff they sell, the association is strongest with vinyl. Among the highest selling artists on the platform, 30% of all the items sold, including all non-musical merch like T-shirts, are vinyl records. Among the lowest selling artist, vinyl was roughly 5%. Downloadable albums and tracks move in the opposite direction, with top-selling artists moving fewer digital goods as they move up the sales rankings, end quote. Here's your history hit of the week from Ars Technica, a deep dive look at Digital Equipment

Corporation, quote, unless you use a Mac, your computer's CPU has its roots in a deck processor, that failed in the market. In the late 1980s and 90s, DEC was looking to change with the times and evolve its Vax line. In 1992, it introduced the Alpha AXP, later shortened to just Alpha, a risk-based processor designed to compete with the other risk chips on the market, such as Sun Microsystems Spark and Hewlett-Packard's PA risk. Deck was getting its lunch eaten by the risk guys, particularly Sun, but also, to a certain extent silicon graphics with their MIPs products, both of which were based on this newfangled risk environment, says John Culver, a CPU historian and operator of the CPU shack. Alpha won't up them all

by being 64 bits at a time when everything else was 32-bit. Bits don't change the processing power, they just change the amount of addressable memory, and back in 1992, no one was worried about the 4-gigabyte theoretical memory limits of 32-bit computers. In that regard, the alpha was way ahead of its time. Everyone was like, why would I need more than four gigabytes of RAM? It was the right thing to do, but it was too early. You see that in a lot of things where they're ahead of their time, and people don't know what to do with it, said Culver. And finally, this is not tech, but from the New York Times, a look at Hollywood's set guru, the guy who designs the sets for a ton of movies you've seen for decades, now including the upcoming Killers of the Flower Moon.

quote. Since the 1970s, Fisk has been one of Hollywood's most sought-after collaborators, legendary among Autor Writer directors for his ability to help them realize their most ambitious projects. He has built boundless, intricately conceived worlds for Terrence Malik, the thin red line, Paul Thomas Anderson, there will be blood, David Lynch, Mohoholland Drive, Alejandro Inoritu, the Revenant, and others. He is the artist that filmmakers hire to bring the American past to the screen at the impossible human scale it once existed. Jack belongs to one of those now rare species of filmmakers who understand film from almost a renaissance-like tradition, in A Route 2 says. He knows photography, nature, architecture, drama. When Scorsese began planning Killers of the Flower Moon,

a lengthy process in which the director radically revised the script from a story centered on the murder investigators to one following the victims, Paul Thomas Anderson told him, you have to get Jack. But Fisk, who is 77, can be notoriously. difficult to entice to a film. Since 1970, he has designed relatively few. At one point, he took nearly 20 years off. When Scorsese approached him, Fisk was excited by the opportunity to collaborate with the director, but also by the chance to excavate a world rarely depicted on screen. The film takes place in a sliver of lost time, one wedged between more familiar depictions of Native Americans in the 19th century, and the well-worn imagery of the roaring 20s. Bringing this moment of cultural

collision back to life would represent as sweeping a challenge as Fisk had ever faced. The story unfolds in about 40 sets, as varied as Masonic Lodges, Osage Funerals, and Federal Courtenrooms spread over a million acres and costing about 15 million of the film's $200 million budget. The sets would represent a kind of culmination of Fiske's career-long obsession with reclaiming the rough contours of American history. More than any one aesthetic vision, he has sought over half a century to scour away the visual cliches that Marr films, seeking beneath them the vivid woodgrain and forgotten colors of the past, end quote. All right, a ton of show news for you.

First up, I will be in San Francisco from Sunday through Wednesday. We're going to try to hold the listener meetup on Monday night at 7 p.m. at Johnny Foley's, which is at 243 O'Farrell Street, right around the corner from Union Square. Again, Monday night at 7 at Johnny Follies, 243 O'Farrell Street. That's also right across the street from where the comes. So if we spill over, I guess we can head to the hotel lobby. Anyway, in order for you to make it easier to recognize me this time, I think I'm going to try to remember to pack and wear my Arsenal jersey, which is red in case you were not aware. So hope to see some of you on Monday night.

Look out for the Arsenal jersey with Saka on the back. I know it is a Monday night, but what can we do? And frankly, NYC, we've never done a listener meetup, so we should plan one of those soon. Next, Monday is a bank holiday here in the U.S., so I will not be doing a show on Monday, but I do have a bonus episode for you this weekend, so I think I'll release that on Sunday night, so it'll be waiting for you on Monday morning. Look for that, another great portfolio profile episode. And if you're attending the AI Engineer Summit, which is the conference I'm in town for, please make yourself known to me. I can't promise I'll be wearing an Arsenal jersey all week, but you'd be surprised how often people recognize. recognize me from my voice. So talk to you on Tuesday unless I talk to you in person first.