← The AI Daily Brief: Artificial Intelligence News and Analysis
The AI Daily Brief: Artificial Intelligence News and Analysis
Did the Super Bowl Make Americans Like AI Any More?
00:26:20
Loading summary
Transcript1 lines
- [00:01]
A
Today on the AI Daily Brief can AI Super bowl ads change what Americans think of AI? Before that in the headlines, the SaaS apocalypse continues. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. Alright friends, quick announcements before we dive in. First of all, thank you to today's sponsors, kpmg, Scrunch, Optimizely and Blitzy. To get an ad free version of the show, go to patreon.com aidaily brief or you can subscribe on Apple Podcasts. If you are interested in sponsoring the show, send us a Note@ SponsorsIDailyBrief AI. AIDailyBrief AI is also where you're going to find links to pretty much everything else going on in this ecosystem of projects, new software, new communities and new initiatives. For those of you who are paying attention to what we were doing over at AIDB New Year's, our 10 week AI new Year's Resolution program, we've got something in a similar vein coming, although a little bit different. I'm not quite ready to share all the details about it, but the gist is if OpenAI is going to agent first work by March 31, my thinking is why shouldn't the rest of us do the same? You can sign up to get the first information about that@aidbtraining.com and I'm excited to share what we've got cooking over there. For now, let's talk about the SASS Apocalypse. Welcome back to the AI Daily Brief Headlines edition. All the daily AI news you need in around five minutes. Last week we talked a little bit about the current state of the markets when it comes to AI. And now a week on from that, the so called SaaS apocalypse is absolutely taking hold as a major investor narrative. Last week 400 billion in market cap was wiped out in software. The iShares software focused ETF IGV closed last week down 8.7%, compounding a 7.4% loss the week prior. Despite a rebound of 3.5% on Friday, the Software Index is dramatically underperforming the Overall S&P 500, at least from a market perspective. The death of software has become the first big AI narrative of the year and investors are scrambling to make sense of it. Many are rushing to explain that SaaS isn't going anywhere and that the notion that large companies will vibe code their own software is preposterous. CNBC featured comments from multiple software CEOs who think the narrative is getting way ahead of reality. Box CEO Aaron Levy argued that the narrative quote somewhat misunderstands this idea of where companies tend to spend their resources and their time and energy. He claims that companies will keep their SaaS subscriptions and look for agents to build on top of them. In other words, he said, that actually could present an opportunity for the SaaS giants to evolve their products with AI. Levy argued that in the 20 year history of cloud software, quote, this is the most exciting moment we've ever had. Salesforce CEO Marc Benioff was singing a similar tune. Salesforce was of course early to the agent theme, with Benioff claiming that AgentForce is the fastest growing product in the history of the company. He also noted that we've got all the customer data, so that moat should remain intact. Still, corporate America is very open to experimentation right now. The Wall Street Journal spoke with Authentic Brands Chief Digital Officer Adam Cronengold about how new AI tools are impacting work for the conglomerate that owns Reebok and Champion. Remarking on the newly released Claude plugin for legal work, which caused one leg down in that big dip last week, Cronengold said the company already has software to handle contract review. But he added, everyone feels very empowered to raise their hand and say, hey, how can we fold this in? You get the impression reading the piece that when it comes to customers, there is a feeling that the friction for experimentation has dropped meaningfully. Brad Gerstner of Altimeter Capital offered an explanation for the dramatic drop in software stocks. Even if you don't believe that SAS is doomed, software has been the best performing sector of the S and P for the past two decades on the back of extremely strong growth. Gerstner argued that forward growth rates are far less clear. Software had been valued at 30 to 35x profit, implying a long horizon view that SaaS had multiple decades of over performance. If AI contracts that horizon to a single decade, that dramatically changes the trajectory for these companies. As Brad put it, AI disruption lowers predictability of future cash flows. Buco Capital summed it up say it with me, we only buy accelerating top lines. Staying on the more nuanced argument theme, one also doesn't have to believe that SaaS companies are empirically doomed to understand that the traditional seat model is looking less and less viable. First of all, even if they keep the seat model, if companies are slashing their headcount, SaaS can't sell as many seats. Eric Goldhar writes, we used to buy software for humans to use. Now we buy agents to do the work. If your product charges by the user, you're selling a tax on productivity. When OpenAI CEO Sam Altman was asked in an interview, is software dead? He somewhat obliquely got at this idea that the fundamental business model of software has changed, saying every company is now an API company, whether they want to be or not. This is most certainly going to be the theme that defines at least the next week in markets, so one will be watching closely. Meanwhile, Thomson Reuters has said that AI is delivering tangible benefits for the legacy legal research firm. The company was one of the hardest hit from last week's release of the legal plugin for Claude coworker. Their stock fell 20% just last week, and despite Thomson Reuters aggressively having added AI features across their product suite, it seems that the market doesn't put too much stock in that. In spite of that, though, CEO Steve Hasker told investors during last week's earnings call, we are seeing tangible benefits from our continued investments in AI. This year they plan to scale up agentic features within the platform in a bid to keep up with offerings from Anthropic and OpenAI. Interestingly, their financials were a perfect demonstration that strong results in the present aren't enough for investors. Thomson Reuters reported 7% revenue growth and 9% growth in EBITDA profits, with margin increasing to 44.3%. Both the AI features in their platforms and content licensing contributed meaningfully to the bottom line. Still, investors are concerned that legacy software platforms will face inevitable disruption from the AI startups. Morningstar maintained their fair value level on the stock, suggesting it's now massively oversold, while Morgan Stanley slashed their price target by a third, writes Finimize. The big question is shifting from will legal gen AI sell? To whether pricing power and renewals hold up as lookalike features spread. When that uncertainty rises, valuation multiples often compress across the software group, even for strong incumbents. Microsoft was another one of the big losers of the software wipeout last week, but they could be positioned for a comeback as agent strategies take hold. The stock lost 6.7%, losing 218 billion in market cap. Now, unlike most of the software stocks, Microsoft faces dual headwinds. In addition to the cracks that are forming in their SaaS business model, Microsoft is also lagging behind the other hyperscalers in their cloud business. Last week's slide and growing competition from the startup labs was enough to trigger a response from Microsoft's top sales executive. According to the Information, Commercial CEO Judson Altoff circulated a memo to sales staff explaining how the company was positioned relative to OpenAI's New Frontier product The memo provided talking points for selling Microsoft's agent management platform, including specific advantages the product has over OpenAI's new platform. Primarily, sales representatives were told to emphasize Microsoft's edge in agent security and compliance. Now, we've been hearing a lot of chatter recently out of Microsoft surrounding the latest agents launched by the frontier startup Labs two weeks ago. For example, the information reported that executives were planning their response to Claude code. This included CEO Satya Nadella discussing his experimentation with the product, as well as setting up OpenClaw and encouraging employees to play with the new tools. Nadella was said to be looking for ways to incorporate the latest agentic features into GitHub and other Microsoft platforms. Now it's clear that Microsoft is feeling the pressure to keep up and are making a bet that they need to keep Agent Security at the center of their sales pitch. Which brings us to our final story for today. Speaking of OpenClaw and security, OpenClaw is partnering with VirusTotal to clean up the skills library. Last week, security researchers revealed that skills published to OpenClaw's clawhub were a cesspit of malware. An audit found that around 400 skills included malicious code. These included some of the most popular, like LinkedIn job application and YouTube thumbnail grabber. Now OpenClaw has partnered with VirusTotal to scan every skill uploaded to their platform and verify their malware free. A blog post announcing the initiative reinforced that this won't be a silver bullet. Writes openclaw virus total scanning won't catch everything. A skill that uses natural language to instruct an agent to do something malicious won't trigger a virus signature. A carefully crafted prompt injection payload won't show up in a threat database. However, they hope the scanning will detect known exploits and suspicious code. One thing that I am quite sure of after having fully gone non technical builder in the last couple of months is that security engineers are going to be more in demand than just about anyone I can imagine. If that's a skill set you have, goodness gracious, is this going to be a good year? For now, that's going to do it for the headlines. Next up, the main episode. Hello friends. If you've been enjoying what we've been discussing on the show, you'll want to check out another podcast that I've had the privilege to host, which is called you can with AI from kpmg. Season one was designed to be a set of real stories from real leaders making AI work in their organizations. And now season two is coming and we're back with even bigger conversations. This show is entirely focused on what it's like to actually drive AI change inside your enterprise and as case studies, expert panels and a lot more practical goodness that I hope will be extremely valuable for you as the listener. Search you can with AI on Apple, Spotify or YouTube and subscribe today. Quick question. When was the last time you actually visited a website to research something? If you're like me, AI pretty much. Does that work for you now? That of course raises a new question for brands. If AI is doing the discovering, researching and deciding who or what is your website really for that shift in user behavior, the rise of AI bots becoming your most important new visitors is what my sponsor Scrunch is taking head on. Scrunch is the AI customer experience platform that helps marketing teams understand how AI agents experience their site. Where they show up in AI answers, where they don't, and what's preventing them from being retrieved, trusted or recommended. And it's not just visibility. Scrunch shows you the content gaps, citation gaps and technical blockers that matter and helps you fix them so your brand is found and chosen in AI Answers. Now for our listeners, Scrunch is providing a free website audit that uncovers how AI sees your site, where there's gaps, and how you're showing up in AI versus the competition. Run your site through it at scrunch.com aidaily Most marketing teams aren't short on ideas, but what they are short on is time. And that's exactly what Optimizely Opal gives you back. With AI agents that handle real marketing workflows. You know, like creating content and checking compliance, generating experiment variations, personalizing user experiences, analyzing pages for geo, even tasks like approvals and reporting. It's your AI agent orchestration platform for marketing in digital teams, plugging seamlessly into the tools you already use, handling the boring busywork and keeping everything on brand that leaves you marketers with more time to do your actual job. See what Opal can automate for your team by signing up for a free Enterprise Agentic AI workshop with Optimizely. Learn more at optimizely.com TheAIDaily Brief with the emergence of AI code generation in 2022, Nvidia Master Inventor and Harvard engineer Sid Pareshi took a contrarian stance. Inference, time, compute and agent orchestration, not pre training, would be the key to unlocking high quality AI driven software development in the enterprise. He believed the real breakthrough wasn't in how fast AI could generate code, but in how deeply it could reason to build enterprise grade applications while the rest of the world focused on co pilots. He architected something fundamentally different. Blitzi the first autonomous software development platform leveraging thousands of agents that is purpose built for enterprise scale code bases. Fortune 500 leaders are unlocking 5x engineering velocity and delivering months of engineering work in a matter of days with Blitzi. Transform the way you develop software. Discover how@blizzi.com that's blitzy.com. Welcome back to the AI Daily Brief. Today we are talking about the AI ads in the Super Bowl. But the reason that this is deserving of an entire main episode is that ultimately this isn't really a show about advertising. Instead, it is a show about public perception and what, if anything, this advertising did to change that? The context coming in is a fairly significant antipathy among the average Americans towards AI. A few weeks ago, the New York Times published a piece called why Do Americans Hate AI? And unfortunately, this isn't just media bluster to run through just some of the recent statistics around American perceptions of AI. According to an Edelman study, only 32% trust it. A Pew study found 59% have little or no confidence that companies will develop AI responsibly. Also from Pew, 52% are more concerned than excited about AI, versus only 10% who are more excited than concerned. Gallup found that 73% expect AI will cause net job loss, and Searchlight found that 51% think AI will replace, not supplement, human work. That, of course, sets up these super bowl ads for a very different task than other advertisers. The Washington Post got at this question with their piece can these super bowl ads make Americans love something that they don't like? Now, of course, there was always going to be some portion of people who are going to write this off as the latest example of a bubble and not actually spend any time engaging with this at all. However, I continue to believe that while Americans are at this point negatively inclined towards AI, for the vast majority, this is not some hardened principled position. This is people rightly understanding that these tools are powerful, and indeed powerful enough that there could be negative implications that they're not sure about. They're commenting from a place of larger economic anxiety that isn't just about AI, but which AI feeds into. They're being asked in a divisive landscape that wants extreme opinions and one which tends to drown out the vast majority of middle voices who are just trying to figure things out. So the question is, how did these ads do with that set of middle voices, not the people determined to hate AI and not the people who are already sold on it, but the group which I consider the vast majority who are still trying to figure out what it's going to mean for them. By Adweek's calculations, 23% of the ads were either for AI companies or highlighted the way that they used AI in their production. So let's go through and see how some of them did and what they might have meant for these broader perception issues. Let's follow up first on Anthropic, given that we covered it last week, summarizing my points for those of you who missed that rant. The ads, for those of you who didn't see it all start with someone using AI to get advice about something that they're trying to achieve or something that they're trying to do, which starts benign enough, but eventually turns into a sales pitch for some product. The tagline, at least the one that was released last week on Social, was Ads are coming to AI, but not to Claude. This was of course, a swipe at OpenAI, who had announced that ads were coming to ChatGPT. Like everyone else, I thought that the ads were well done and that they were funny, if you have the proper context. My critiques of them were two part. First, from an effective advertising perspective, while these killed on Social, for people who were enfranchised enough to actually know that OpenAI had announced that ads were coming, I thought that they were going to be incredibly confusing for the average person who had no idea it would be different if for the last three months people have been complaining about ads in ChatGPT. But that's just not the case, which led to my second complaint, because people wouldn't have that context. It felt to me like the vast majority of people who did enjoy this were going to be those who interpreted it as a critique of AI in general. Think about what the ad says. If you don't know that it's trying to go after ChatGPT, it's people with genuine concerns and challenges in their life, but vulnerably turning to AI only for it to turn around and try to sell you something. The point it's making, then, at least to that perception, reinforces the idea that AI is just the latest thing from tech billionaires to separate you from your money and time. So how did they land? The short answer is not that well. From Ad Week, early audience response suggests the message struggled to land. According to an I spot survey of 500 viewers, the ADSlikability score placed it in the bottom 3% compared with Super bowl ads over the past five years. Its top two box purchase, intent, scored 24% below Super bowl norms and 19% below ads in its category that aired over the last 90 days. Viewers most commonly described their reactions as WTF, signaling confusion around both the message and the execution. So there you have it. Likability in the bottom 3% compared with Super bowl ads over the last five years is not something to be proud of. Now. One cool thing from Claude was that they added live sports scores and stats into the app. Now this is the type of feature that 100% is not for me, but to the extent that you were trying to turn your chatbot into a portal to connect to the rest of the Internet, this for some users could be really valuable, rob Haysfield said. Wait, why didn't Anthropic put this in their super bowl ad? Tech publication TVPN tried to have some fun with it, running a spoof where they offered Claude with ads. Now we also talked last week about how OpenAI had bit back pretty hard on social, calling Anthropics ad all sorts of dishonest, but they hadn't yet revealed their super bowl ad. The tagline of this ad was you can just build things. The ad focuses on the hands of the subject and has a bunch of smash cuts to short vignettes of people doing stuff, making a marble track, building electronics, answering in class, installing Linux, creating a pigment for painting, playing chess, working on robotics, getting their Apple on. They flashed through luminary geniuses like Alan Turing, Grace Hopper and Albert Einstein. The ad was not just Codex coded, it was Codex branded OpenAI CMO Kate Rauch wrote. We're living through a time when people can build things that were previously out of reach. The message is around participation, agency and leaning in to use these tools to do things you just couldn't before reinforcing the reframe that we've been living through and documenting on this show, she added. Now AI is not only answering your questions, but is going to do things on your behalf. Rauch described Codex as a sign of material change underway in the industry. Now the funny thing is, I think that if you had shown AI people this ad a couple of months ago without the ending, a majority of them would have bet it was for Claude code rather than for Codex. But that's not where it landed. And so the question is, how did people receive it? Ad Age basically said it was OpenAI trying to position ChatGPT as the Kleenex of AI. In other words, ChatGPT just being synonymous with AI, so taking advantage of the brand positioning they already had to make kind of a more highfalutin ad. Generally, the response I saw to it was fairly positive, if benign. I saw it on some lists as the top ranked of the AI ADs, although I think the highest placement I saw was maybe the New York Times list had it at 17th best out of the total 66. For others it was fine but forgettable, but ultimately a positive vision of the future. Now one crazy wrinkle on everything was that after the ad came out, this post popped up on Reddit that was purportedly from a disaffected OpenAI staffer who said that this ad was actually a last minute replacement for an ad that would have teased OpenAI's device. At first they showed just a screenshot of what looked like Alexander Skarsgrd looking at a weird tabular thing, and then a few minutes later, an entire 30 second ad appeared. It indeed appeared to be Alexander Skarsgrd with some weird chrome ear things, pressing a shiny chrome device and looking confusedly at it. Reddit founder Alexis Ohanian said the ad is beautiful. They should have run this one. But apparently it was all just an elaborate hoax. The Verge writes. Whoever was behind this hoax had been working on it for some time and approached spreading their story on multiple fronts. Analyst Max Weinbach got messages about it a couple of weeks ago and Ad Age reporter Gillian Follett wrote, a fake headline is circulating on X that falsely claims to be my reporting. I did not write that story and neither I nor Ad Age has published a piece saying OpenAI changed its Super bowl ad. OpenAI CMO Kate Roesch again responded, fake headline. An entire fake website claiming we changed our super bowl ad at the last minute. Someone going to a lot of trouble. Honestly, I do not know what to make of this one and what the incentive for someone to create this entire ad and backing campaign was. Maybe we'll learn over time, but for now just a weird wrinkle in this overall story. Moving on, Microsoft and Meta both tried to tailor their AI pitch to the sports crowd. Microsoft's ad actually wasn't specifically produced for the super bowl, but was part of a larger campaign around Copilot, and it's an ad all about how Copilot helped NFL teams and players make better decisions through its ability to process lots of data. Fairly unremarkable, but at least on brand with the setting. Meta's ad focused on the Oakley version of their smart glasses. It opens on Marshawn lynch asking the AI to play his Beast Mode playlist then progresses through a series of users making other requests of the assistant that included Spike lee and streamer iShowSpeed. Interestingly, Fortune suggested that this ad wasn't really aimed at the consumer. Kimberly Whitler, an associate professor at the University of Virginia, said, they're not just communicating to consumers, they're also communicating to investors who watch these ads. They're reinforcing this kind of view that the company is very innovative. I don't know, man. I think that Meta has recognized that although certain parts of their AI strategy haven't gone so well, they are in a unique position when it comes to these wearables and that there's a lot of use cases for AI glasses like the ones they show here, where the glasses just do things for you as you're living your life at high speed that might appeal to regular people. I think the ad does a good job playing to Meta's strengths in this area and fitting in with the larger super bowl vibe. Now, the AI ad that had the most acclaim was the ad from Google focusing on Gemini. In it, it shows a mom and a son getting excited about their new house. They're taking images of the raw potential and imagining what it could be in the future. Thanks to the help of Gemini. Google is historically one of the few super bowl advertisers that's actually daring enough to try to move out of the tried and true patterns of funny to pull on the heartstrings. And they've gotten pretty good at it. It's a sweet ad and I guarantee that if you're a parent, it would have made you feel something. Because the point is, it's not really about AI. It's about how AI can be a part of the crazy journey of life. Artificial Nightmares wrote Google really killed it with their super bowl ad. Clear, concise, human, and it instantly showed you the value of their product and why you need it. With a real world example, Every's Dan Shipper wrote, low key Google had the best AI super bowl commercial for consumers. Actually good at pitching the product and explaining why you might use it. Now, one company that went in the exact opposite direction and decided to play on and simultaneously validate and also normalize and make fun of fears of AI taking over was Amazon's Alexa ad with Chris Hemsworth in it. Chris Hemsworth comes in to see a new Alexa in his house, only to freak out and describe all the ways that it might try to kill him, which are then of course hilariously documented in the rest of the ad. Endured on Twitter, writes, honestly, brilliant marketing. Everyone else is trying to sound safe and serious. Amazon just goes, we know what you're thinking. Now my guess is this one is going to be pretty divisive with some people not appreciating Amazon making light of a fear that actual people have. But I think running all the way in this opposite direction is actually a strategy with some merits and opened up a lot of really funny territory for them. Now what's interesting is that it wasn't just the big guys running ads as well. Genspark apparently decided around Christmas to do an ad and their agency put it together, of course, with the help of AI in just five weeks. The ad featured historic slacker Matthew Broderick, subtly making a nod to his Ferris Bueller character, walking through an office and explaining how genspark could give everyone the day off. It makes slides, it does some Excel work with the idea to make clear AI's time saving benefits. I saw a couple reputable lists think it was the most effective of the AI ads. But there are also a lot of people who felt like Stephen Breach on Twitter, who wrote, this is pretty much telling our employers they don't need us anymore. Question mark, question mark, question mark. The genspark social account responded saying, get the concern, but it's actually the opposite. AI doesn't replace us. It handles the grunt work so we can focus on the creative, strategic stuff that actually matters. Even better, it gives us time to build something of our own. That's the future we're excited about. Stephen, for his part, wasn't buying it, saying, this sounds like the AI response now. Vibe coding toolbase44, now part of Wix, came at the same theme of you can just build stuff that OpenAI did, but in a much more direct way. In their ad, they show people in an office discovering delightedly that they can build apps that weren't possible before. For my money, this was easily the most underrated of the AI ads. First of all, unlike the genspark ad, it's not showing AI doing people's core work. It's showing how they can have new capabilities that weren't possible before. In that way, it's inherently empowering. Second, the ad is just straight funny. The stuff that they build creates a lot of room for humor. The tagline it's app 2 you lands. And I just think it was an actually good spot. Now we're getting along, but there are just two more that we need to talk about. The first is from Svedka, who proudly declared that they had made the first fully AI generated Super bowl ad ever and people were not super impressed, to put it mildly. The ad features honestly the only way to put it is horror movie robots chugging, Svetka dancing at a rave with the tagline shake your bots off. AI filmmaker PJ Ace wrote, I would not be proud of this lol. In fairness we're talking about it but God it sucks. That I think is fairly consensus view so I'll just move on to our last spot which is from AI.com now AI.com didn't exist until very recently. The origin story is that the founder of crypto.com spent $70 million to buy the domain, making it, I believe, the most expensive domain purchase of all time. The ad sent people to AI.com to claim their handle and launch their own AI assistant, which enough people did that it crashed their website. This was the icing on the bubble cake. Michael Podlowski writes, crypto guys, no AI background. Buy AI.com for 70 million. Burn 10 million on a super bowl ad slogan accelerating the arrival of AGI Ask you for a credit card right away just to claim a handle. Website looks like a cheap 5 coded mess crashes instantly with a 5:04 gateway timeout. This looks like the absolute peak of the AI bubble. And fine, may be so, but but the part that I most took notice of was this line. Turns out it's just a thin Open Claw wrapper. And Indeed founder and CEO Chris on Twitter in his post AI.com is now live in beta, writes AI.com is the world's first easy to use and secure implementation of OpenClaw, the open source agent framework that went viral two weeks ago. We made it easy to use without any technical skills while hardening security to keep your data safe, even in a world that moves fast. The fact that Open Claw went from non existent to part of a Super bowl ad in three weeks has got to be some kind of record. So ultimately, could these super bowl ads make Americans love something they don't like? Honestly, I'm not really sure. There were highlights and lowlights, but I guess overall it could have been a lot worse. So maybe we'll call that a win for now. That's gonna do it for today's AI Daily Brief. Appreciate you listening or watching as always. And until next time, peace.