The AI Daily Brief: Artificial Intelligence News and Analysis - OpenAI Debates Advertising

Episode Date: December 4, 2024

OpenAI's monetization approach may shift as discussions about introducing advertising gain traction. Reports highlight the financial challenges of running advanced AI models and the potential role ads could play in funding operations. This video explores the implications of ads on trust, user experience, and the future of generative AI. Are ads inevitable, or is there another way forward? Brought to you by: Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlw Plumb - AI automation that just works - https://useplumb.com/ The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614 Subscribe to the newsletter: https://aidailybrief.beehiiv.com/ Join our Discord: https://bit.ly/aibreakdown

Today on the AI Daily Brief, will advertising be the business model for AI as well? And before that, in the headlines, an AI casualty over at Intel. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. To join the conversation, follow the Discord link in our show notes. Welcome back to the AI Daily Brief Headlines edition, all the daily AI news you need in around five minutes. We kick off today with a story of a so far unsuccessful AI turnaround. Intel CEO Pat Gelsinger has abruptly stepped down. from the troubled U.S. chipmaker.

According to Bloomberg reporting, the shock announcement came after a board meeting to receive a progress report on plans to win back market share from Nvidia. Apparently, that didn't go so well, with sources stating that Gelsinger was given the option to retire or be forced out. There is no replacement in mind with the firm CFO and executive VP sharing the role while the board searches for a new CEO. Gelsinger was heralded as the company's savior when he was installed in 2021. He worked for the company for over three decades, starting as a teenager,

in 1979. He's credited as the chief architect of the Seminole 486 processor and spending a decade as the company's CTO. Gelsinger was central to the golden age of Intel. And that was the vision he presented a little over three years ago, a return to Intel's glory years. His plan was to expand the company's network of chip foundries and reestablish Intel as the world's largest chip maker. This included a massive new complex in Ohio and upgrades to their Arizona, New Mexico, and Oregon facilities. To carry out this grand vision, the company was earmarked for $8 billion in the Chips Act and positioned as a national champion for U.S. chip making. Intel is viewed as a strategically important company as it's one of the few chip manufacturers that could credibly compete with TSMC.

Nvidia is also a critical U.S. company in the sector, but they don't operate their own fab so can't easily substitute. Gelsinger's plans fell apart almost immediately. PC sales declined sharply in the latter half of 2021, as work from home arrangements were wound back. The company then lost valuable customers in Waymo and Sony, failing to win the contract to manufacturer of PlayStation 5 processors. None of the new U.S. fabs are up and running, and the foundry business has been restructured into an independent business. This was celebrated by shareholders as a way to enable the company to seek external customers, but it was also a sign that Intel wasn't seeing enough chip demand to justify the fabs they already had, let alone build new ones.

Interviews made it sound like Gelsinger was getting nowhere fast. In a 2022 interview with the Verge, he said, not only do we have to rebuild the internal culture and execution, but we also have to rebuild our customers' confidence. The company reported 7 billion in operating losses last year and laid off 15,000 workers in August as part of a $10 billion cost-cutting exercise. With his failure to deliver now acknowledged, Gelsinger said in a press release, leading intel has been the honor of my lifetime. This group of people is among the best and brightest in the business, and I'm honored to call each and everyone a colleague. Today is, of course, bittersweet as this company has been my life for the bulk of my working career. I can look back with pride at all that we have accomplished

together. It has been a challenging year for all of us as we made tough but necessary decisions to position Intel for the current market dynamics. And from here, it's a big question about whether Intel has a path to recovery. Its latest round of consumer processors were met with a lukewarm reception, failing to deliver meaningful improvements. Tests of their next generation 18A manufacturing process have hit problems with uneconomic yields and won't be in full production until 2026. With the departure of Lipbutan in August, the company now has zero board members with semiconductor experience. Qualcomm approached the company for an acquisition in September, and it seems increasingly likely that may be the way the Intel story ends. Frank Yeri, the chairman of

the board said on Monday, with new leadership, we will continue to act with urgency on our priorities, simplifying and strengthening our product portfolio and advancing our manufacturing and foundry capabilities while optimizing our operating expenses and capital. We are working to create a leaner, simpler, more agile intel. In other words, the exact same goals, aside from perhaps slimming down the company, that Gelsinger brought to the table in 2021. The stock chart probably says it all. A 5% jump as the news broke with the market recognizing that a much-needed change was happening, then a full collapse to close the day lower in recognition that a simple CEO swap

probably won't fix what's broken. Next up, another story in the chip space, Jeff Bezos has joined a massive bet on Challenger chip-making startup Tensterent. The company has closed almost 700 million in Series D funding that would value the company at $2.6 billion. The round was led by Samsung Secure. NafW partners with participation from Hyundai and Healthcare of Ontario Pension Plan. The goal of the startup is simple, if ambitious, to take market share from NVIDIA. The company is led by Jim Keller, formerly AMD's chip design guru. He was the lead architect of the AMD K-8, which underpinned the Athlon 64,

the first 64-bit processor available for consumer-grade hardware. Keller was also involved in designing the Apple A4 and A5 processors alongside many other AMD architectures. The big bet seems to be that Keller can deliver a revolutionary AI architecture, architecture for cutting-edge two nanometer chips expected to come to market by 2027. In the shorter term, the company is focused on solving for power use and cost efficiency. They are building largely on open source and more commonplace technology, allowing them to use cheaper components and offer greater interpretability than Nvidia. Keller said, in the past, I worked with proprietary tech and it was really tough.

Open source helps you build a bigger platform. It attracts engineers. And yes, it's a little bit of a passion project. In contrast, of course, Nvidia produces their own proprietary ecosystem to completely surround their AI chips. From interconnects and data center layouts to the software used to train models, Nvidia controls their entire infrastructure stack. TenStorant plans to release a new AI processor every two years, which is half the release speed of Nvidia. The company, which was founded back in 2016, will produce their first mass production run next year using TSM and Samsung Fabs. To date, the company has 150 million in customer contracts. And Keller seems pretty focused on getting

going, tweeting, round D closed. 693 million. Many thanks to our investors, lots to do. Let's get to work. Lastly, today, something which I have a feeling we are going to get way more into in the future. World Labs has unveiled their first project, an AI world generator. The startup founded by Fei-Fei Lee, the so-called godmother of AI, has demoed a system that can generate a 3D world based on a single image. They're calling the tech spatial intelligence, but other segments of the industry refer to the emerging category as world models. In a blog post, World Labs wrote, this lets you step into any image and explore it in 3D, beyond the input image, all is generated. We've already seen a bunch of Gen AI products that can create video and static 3D environments from reference images,

but this product creates scenes that are fully interactive and modifiable. One example is a scene of a medieval village. The user is in full control and can walk down the alleyways that weren't visible in the source image. There's also a very cool demo where the user can stroll the streets of Paris as depicted in Cafe Terrace at night by Van Gogh. World Labs wrote, Most generative AI tools make 2D content like images or videos. Generating in 3D instead improves control and consistency. This will change how we make movies, game simulators, and other digital manifestations of our physical world.

Like I said, there is a lot here to come back to. There are some who think that this is a better path to reaching AGI than just training on a bunch of text data. And so I'm sure we will be coming back to this. But for now, check out worldlabs.a.i. And you can actually walk around and demo some of these experiences. For now that that is going to do it for today's AI Daily Brief Headlines edition. Next up, the main episode. Today's episode is brought to you by Rocket Money.

We are coming up on the beginning of the new year, and that is a perfect time to get organized, set goals, prioritize what matters most, which for many of us is going to be financial wellness. Thanks to Rocket Money, those goals, especially around money, feel achievable. Rocket Money shows you all of your subscriptions right in one place, helping you easily cancel those that you've maybe forgotten that you're actually paying for. Rocket Money also pulls together all of your spending across your different accounts so that you can clearly track spending habits and see where you can cut back. Rocket Money is a personal finance app that helps find and cancel unwanted subscriptions,

monitors your spending, and helps lower your bills so you can grow your savings. Their dashboard gives you a clear view of your expenses across all of your accounts. You can easily create a personalized budget with custom categories. You can see your monthly spending trends in each category to know exactly where your money is going. Rocket Money will even try to negotiate lower bills for you. They automatically scan your bills to find opportunities to save, and then you can ask them to negotiate. They'll deal with customer service so that you don't have to. Rocket Money has over 5 million users and has saved a total of 500 million in canceled subscriptions,

saving members up to $740 a year when using all of the app's premium features. Cancel your unwanted subscriptions and reach your financial goals faster with Rocket Money. Go to rocketmoney.com slash AI breakdown today. That's rocketmoney.com slash AI breakdown. Today's episode is brought to you by Plum. Want to use AI to automate your work but don't know where to start? Plum lets you create AI workflows by simply describing what you want. No coding or API keys required. Imagine typing out AI, analyze my Zoom meetings and send me your insights in Notion,

and watching it come to life before your eyes. Whether you're an operations leader, marketer, or even a non-technical founder, Plum gives you the power of AI without the technical hassle. Get instant access to top models like GPT40, Claude Sonnet 3.5, assembly AI, and many more. Don't let technology hold you back. Check out Use Plum, that's Plum with a B, for early. access to the future of workflow automation. Today's episode is brought to you by Vanta.

Whether you're starting or scaling your company's security program, demonstrating top-notch security practices, and establishing trust is more important than ever. Venta automates compliance for ISO-2701, SOC2, GDPR, and leading AI frameworks like ISO-402, and NIST AI Risk Management Framework, saving you time and money while helping you build customer trust. Plus, you can streamline security reviews by automating questionnaires and demonstrating your security posture with a customer-facing trust center all powered by Vanta AI. Over 8,000 global companies like Langchain, Lila AI, and factory AI use Vanta to demonstrate

AI trust and prove security in real time. Learn more at vanta.com slash NLW. That's vanta.com slash NLW. Today's episode is brought to you by Super Intelligent. Every single business workflow and function is being remade and reimagined with artificial intelligence. There is a huge challenge, however, of going from the potential of AI to actually capturing that value. And that gap is what Super Intelligence is dedicated to filling. Super Intelligence accelerates AI adoption and engagement to help teams actually use AI to increase productivity and drive business value. An interactive AI use case registry gives your company full visibility into how people

are using artificial intelligence right now. Pair that with capabilities building content in the form of tutorials, learning paths, and a use case library. And Super Intelligent helps people inside your company. show how they're getting value out of AI, while providing resources for people to put that inspiration into action. The next three teams that sign up with 100 or more seats are going to get free embedded consulting. That's a process by which our super intelligent team sits with your organization, figures out the specific use cases that matter most to you, and helps actually

ensure support for adoption of those use cases to drive real value. Go to Bsuper.a.i to learn more about this AI enablement network. And now back to the show. One of the big questions lurking around the Gen AIS space is who pays for this stuff? What's the business model? And this is obviously made all the more challenging by the fact that as opposed to some other categories of software, there are real marginal and incremental costs for each usage of a big foundation model like ChachyPT. Now, of course, this is why we've gotten a Gen AI industry that charges for its software right out of the gate. In fact, I think there's been in some ways a retraining of people to pay for the software that they use,

which I think is a generally healthy thing. It's certainly different than Zerpera companies, where everything was basically subsidized by VCs, leading eventually to business model chaos as things moved into public markets, and in general we moved off of zero interest rates, and the availability of risk capital started to decrease. But none of that answers the question ultimately of how AI is ultimately going to make money. And maybe one of the most natural questions is whether these platforms will have to incorporate advertising alongside their subscription revenue. Some players have already walked down this path. Perplexity three weeks ago published a blog post

why we're experimenting with advertising. They write, to fully deliver on our mission to spark the world's curiosity, we need to invest in building not just a beloved product, but a robust and self-sustaining business. That's why starting this week, we will begin experimenting with ads on perplexity. From there, they get into how they'll draw the lines, what their particular ad product will be, but the point for the purpose of our conversation today is that it's clear that perplexity does not believe that subscriptions alone can get them where they need to be. But what about the real biggies? Well, in an interview with the Financial Times, OpenAI CFO Sarah Fryer said that the company was weighing ads as a revenue model. She said they plan to be,

quote, thoughtful about when and where we implement them. Curiously, in a statement after the fact, she walked the comments back slightly, adding, our current business is experiencing rapid growth and we see significant opportunities within our existing business. While we're open to exploring other revenue streams in the future, we have no active plans to pursue advertising. And indeed, even in the interview before that statement, Friar had made it seem as though there were more pressing revenue considerations. She said, I don't preclude ads,

but for now there's lots of low-hanging fruit in the way we're doing things. Still, FT sources suggest that the plan is a little more active than Friar might be willing to admit. They claim that OpenAI has recently been hiring advertising talent from big tech rivals, including meta and Google, supported by FT's review of LinkedIn profiles. The company also has Kevin Wheel as their chief product officer, bringing his experience building out ad revenue at X and Instagram. Friar said the good news with Kevin Wheel at the Wheel with product is that he came from Instagram. He knows how introducing ads works. And yet, Sam Altman has been vocally against the idea. During a fireside chat last month at the Harvard Business School,

Altman said that ads would be a, quote, last resort. I'm not saying OpenAI would never consider ads, but I don't like them in general. And I think that ads plus AI is sort of uniquely unsettling to me. The FT reported, however, that Altman is warming up to the idea, according to a person familiar with his thinking. Still, he would have to be really warming up to move off his previously very strong positions. Earlier this year, in an interview, Altman called advertising a, quote, momentary industry. He said, I kind of hate ads just as an aesthetic choice. I think ads needed to happen on the internet for a bunch of reasons to get it going, but it's a momentary industry. The world is richer now. I like that people pay for ChatGPT and know that the answers they're getting

are not influenced by advertisers. I'm sure there's an ad unit that makes sense for LLMs, and I'm sure there's a way to participate in the transaction stream in an unbiased way that is okay to do, but it's also easy to think about the dystopic visions of the future where you ask ChatGPT something, and it says, oh, you should think about buying this product, or you should think about going here for your vacation or whatever. And I don't know, we have a very simple business model, and I like it. And I know that I'm not the product. I know I'm paying and that's how the business model works. And when I go use Twitter or Facebook or Google or any other great product, but ad-supported great product, I don't love that. And I think that gets worse, not better, in a world with AI.

Still, one of the issues facing Open AI is that the numbers are really hard. The company is on track to burn $5 billion in cash this year, which is set to escalate as training costs ramp up for the next generation of frontier models. That means the company is locked into achieving record fundraising rounds each year for the foreseeable future. The question is whether the revenue model of charging for subscriptions in API access can grow large enough to address the challenge. An FT source said they are pursuing consumer activity and consumer search. The API is not a high-margin business. Christopher Penn, chief data scientist at Trust Insights.A.I. An AI firm for marketer said, it's long been known that OpenAI is burning cash at a crazy rate in order to keep up their

operations. When you think about it, chat GPT as a tool is absurdly low price for what it delivers. As they've debuted new models such as the flagship 01 model, it's clear from the cost tokens in that model that it is a crazy expensive model for them to run. The hardware requirements and the processing power point towards the model being the future of the company, but at a substantially increased price. Advertising allows someone else to pay the bill other than the users. Penn also commented, consumers generally don't have a problem with ads as long as the ads are not deceptive and they're clearly marked. If OpenAI in the chat GPT interface has a response and there's a big bright yellow, here's an ad block that clearly denotes where the generated

answer and the advertising placed answer are, that would be good. If on the other hand, the user can't tell the difference between a purely generated answer and an advertising place answer, that would substantially impact trust because you don't know whether the answer is actually correct or was paid to be correct. In today's media environment where trust is at all-time lows for everything, that could be substantially harmful to their business. Craig L.M. Eli, the chief creative officer at Code and Theory, has a different view, stating, if OpenAI does include ads, it's important that they don't feel like ads. They have to improve the experience, not just disinterrupt it. He argues that OpenAI should, quote, consider rethinking ads altogether,

blending them into what utility users have come to expect. He urged OpenAI to embrace tools or experiences that serve the user, such as, quote, curated recommendations that feel anticipatory, personal, helpful, and aligned with the prompt. In his view, the magic is in making ads that feel like they belong and are useful and smart. VC. Eric Frankie writes, of course OpenAI is exploring an ads business. ChatGPT will have one billion users soon. API and subscription revenue, currently $4 billion, will tap slow eventually.

Cost to build and maintain the model currently $5 billion will continually increase. Properly executed ads will be scalable, high margin, and additive to the user experience. David Bressler writes, OpenAI will make more money from ads than anything else. OpenAI hired the person who ran Google Search Ads team, then they added the ability to search the internet, then they added UT source equals chatGBT.com to all outbound links, allowing sites to see attributed traffic and conversions. In the same way, we're heavily dependent on chat GPT for productivity, there will be a world where we're dependent on chat GPT for web traffic in the same way companies are currently dependent

on paid ads. Others think it's going to be a little bit harder than it seems. Sasha Koleski writes, chat GPT's ad platform will not be an easy sell to advertisers. Google search makes most of its revenue from low funnel complex services requiring clicks. Insurance, accident lawyer, JFK car rental. Advertisers pay very high CPCs for this. OpenAI's volume is generally upper funnel Q&A, and advertisers will not pay very high CPCs or likely CPMs for this because users are not looking to make purchasing decisions. So in my opinion, advertisers will see OpenAI's platform for the time being at least as the low intent of display ads with the low volume of search. Beyond the novelty, without a change in chat GPT user behavior, this is not super

exciting. Others just really don't want this. Parrot writes, OpenAI exploring ads send shivers down my spine, personalized ads at its worst. Adam Butler writes, I'm currently in favor of very light touch AI regulation, but if Open AI embraces an ad model, I'd immediately pivot to advocating for the harshest possible regulations and unsubscribe. Now, I tend to think this is probably an inevitability. It's been clear for some time to me that part of what ChatGBTGPT is competing for is not just this new category of LLMs, but for the place of your gateway to the entire world of information.

In that way, they are directly competitive with Google Search. And I think eventually there will be more and more intent that can be inferred from chatGB prompts and queries that lead inevitably to the path of ads. The more interesting question to me in the short term is not whether it's a good idea from a business standpoint, but how much in general the business side of these companies is distracting them from their mission to reach AGI. One of the more notable events I think of the last year was Ilya Sutskever leaving Open AI to found the company that he calls Safe Superintelligence, which is willfully unwilling to consider business models before AGI. They wrote,

We have started the world's first straight shot superintelligence lab with one goal and one product, a safe superintelligence. As I is our mission, our name, and our entire product roadman. because it is our sole focus. Our singular focus means no distraction by management overhead or product cycles, and our business model means safety, security, and progress are all insulated from short-term commercial pressures. And SSI does have resources to bring to this mission. They've raised a billion dollars, and so have some capital to burn. But it is a really interesting tension inside these companies around how much they should be focused on their business models versus focused on research and trying to push out the state of the art. I think it'll be fascinated to watch

this question as it comes to bear in the next year. For now, though, if you're looking at hints for what's next for OpenAI, there seems to be a lot of buzz on Twitter after Roon tweeted OpenAI is unbelievably back, and another set of Open AI folks added their own cryptic messages. So will we see something important over the next couple weeks? I will certainly let you know as soon as I know. For now that, that's going to do it for the AI Daily Brief. Appreciate you listening or watching, as always. Until next time, peace.