Ads Come To ChatGPT

The world moves fast. Your workday? Even faster. Pitching products, drafting reports, analyzing data. Microsoft 365 Copilot is your AI assistant for work. Built into Word, Excel. PowerPoint, and other Microsoft 365 apps you use, helping you quickly write, analyze, create, and summarize. So you can cut through clutter and clear a path to your best work. Learn more at Microsoft.com slash N365 Copilot.

Welcome to the Tech Brew ride home from Monday, January 19th, 2026. I'm Brian McCullough. Today ads are finally coming to ChatGPT. Why everyone online can't stop talking about going on clodbenders. Elon's case against OpenAI moves forward again, the Thinking Machine saga roils on again. Big movie about AI everyone is apparently watching. Here's what you missed today in the world of tech.

There's a lot of chatter about using AI agents right now, but no one seems to be talking about how these agents aren't always perfect. Sometimes they delete the wrong files or make changes you didn't authorize, and when that happens you're left to clean up the mess. That is, unless you're using Rubrik Agent Cloud. It's the only platform that allows you to monitor, govern, and rewind AI agent actions. And if your business relies on AI agents, you need those abilities.

On a singular platform, it helps you unleash more agents faster while mitigating risk. If you're running AI agents and want to sleep better at night, Rubrik is worth checking out. Right now, my listeners get exclusive early access to the platform. Head to rubric.com. That's r-u-b-r-ik.com.

It is official. OpenAI plans to test ads below ChatGPT replies for users of the free and go tiers in the US. A source says OpenAI expects to make in the low billions of dollars from ads in twenty twenty six alone. Quoting the FT, the San Francisco headquartered company announced on Friday that it would begin testing adverts on its free chatbot and cheapest paid offering. The ads will appear at the bottom of Chat GPT answers in the coming weeks.

OpenAI said the marketing messages will be clearly labelled and will appear if relevant to the query. The company led by Chief Executive Sam Altman is exploring new revenue streams that could help fund spending commitments of roughly one point four trillion dollars. on computing resources over the next decade. OpenAI anticipates it will make low billions of dollars from advertising in twenty twenty six and more each year thereafter, according to a person close to the company.

In recent months, OpenAI has been enhancing the ability of ChatGPT to store information from conversations that can later be retrieved to provide more personalized responses. The focus on memories provides a powerful tranche of preferences which the company anticipates drawing on to tailor hyper personalized advertising according to the person close to OpenAI. The company said personalization could be turned off and memories deleted.

Earlier this week, Google launched personalized advertising into its AI mode in its search engine, which delivers bespoke discounts to users shopping on the platform. OpenAI is also working on other alternative business lines, including a suite of AI hardware devices, developing bespoke products for governments and businesses, and new shopping tools.

It is also in early talks with investors regarding a new funding round that could raise about eighty billion dollars, according to people familiar with the deal, end quote. A slightly different angle now from Benzinga, which I'm sure is related to what we just spoke about. On Sunday, OpenAI said that its annualized revenue run rate exceeded a$20 billion in 2025, a 233% increase.

from twenty twenty four, accelerating sharply from the prior year's growth when revenue rose from two billion dollars in twenty twenty-three to six billion dollars in twenty twenty-four. This is never before seen growth at such scale, CFO Sarah Fryer said in a blog post. We firmly believe that more compute in these periods would have led to faster customer adoption and monetization, end quote.

The company acknowledged that revenue growth has moved almost in lockstep with its expansion in computing power, highlighting the infrastructure-heavy nature of generative AI. OpenAI said it increased compute capacity from point two gigawatts in twenty twenty three to point six gigawatts in twenty twenty four, reaching roughly one point nine gigawatts in twenty twenty five, nearly a tenfold increase in two years.

Revenue followed a similar trajectory, climbing from two billion dollars to more than twenty billion dollars over the same period. Compute grew three X year over year or nine point five X from twenty twenty three to twenty twenty five, the company said, adding, while revenue followed the same curve growing three X year over year or ten X from twenty twenty three to twenty twenty five, end quote.

The company is reportedly burning more than$17 billion annually, and revenue from subscriptions alone may fall short of supporting its highly compute-intensive AI operations. In December, it was reported that OpenAI was seeking to raise a hundred billion dollars at a valuation of eight hundred and thirty billion dollars, largely to fund further compute expansion, end quote.

But all of the discourse online is about Clawed Code and that Claude Work thing. If you listen to today's Odd Lots podcast episode, it's all about this, though I've not had the chance to listen to that episode yet. Quoting the journal.

They call it getting clawed pilled. It's the moment software engineers, executives, and investors turn their work over to anthropics clawed AI and then witness a thinking machine of shocking capability, even in an age awash in powerful artificial intelligence tools. Many coders spent their holiday breaks on a so-called Claude Bender, testing out the capabilities of the latest anthropic model, Claude Opus four point five, which they used within a desktop coding tool called Claude Code.

Tech companies have been incorporating code writing AI into their workflows for years, and prior models were often compared with a junior software developer. The buzz around Claude's latest incarnation is something different. is Chief Technology Officer at Vercell, which helps develop and host websites and apps for users of Claude Code and other such tools. He said he used the tool to finish a complex project in about a week that would have taken him about a year without AI.

Ubel spent ten hours a day on his vacation building new software and said each run gave him an endorphin rush akin to playing a Vegas slot machine. The clawed zeal has spread widely this month, even to non engineers. Many took to social media to describe the process of building their first software program without ever having learned to code. And despite the code in the name, people are using Claude Code for everything from health data analysis to expense report compiling as well.

Some described a feeling of awe followed by sadness at the realization that the program could easily replicate expertise they had built up over an entire career. It's amazing and it's also scary, said Andrew Duca, chief executive of Awakened Tax, a cryptocurrency tax platform. Duca has been coding since he was in middle school. I spent my whole life developing this skill, and it's literally one-shotted by Claude Code, end quote.

Claude's total web audience more than doubled in December from the previous year, and its daily unique visitors on desktop are up twelve percent globally year to date compared with last month, according to market intelligence companies SimilarWeb and Sensor Tower respectively.

The question for many is what happens next? The bigger story here is going to be when this goes beyond software engineering, said David Sue, chief executive of Retool, a business AI startup. Software engineers make up a tiny fraction of the US labor force, but How far does it go?

This is from Bloomberg. The release of a new artificial intelligence tool from Startup Anthropic on January 12th rekindled fears about disruption that weighed on software makers in 2025. TurboTax owner Intuit tumbled sixteen percent last week, its worst since twenty twenty two, while Adobe and Salesforce, which makes customer relationship management software, both sank more than eleven percent.

All told, a group of software as a service stocks tracked by Morgan Stanley is down fifteen percent so far this year, following a drop of eleven percent in twenty twenty five. It's the worst start to a year since twenty twenty two, according to data compiled by Bloomberg.

The anthropic news we got underlines how difficult it is to assess what growth can look like going forward, said Brian Wong, portfolio manager at Osterweiss Capital Management, which has seven point nine billion dollars in assets. The pace of change is about as fast as I can remember, and that makes things about as uncertain as I can remember.

Anthropics Claude Cowork Service, released as a research preview, can create a spreadsheet from screenshots or produce a draft report from an assortment of notes. According to the company. It was developed quickly, largely with AI. While unproven, the tool represents just the type of capabilities that investors have been fearing and reinforces bearish positions that are looking increasingly entrenched. According to Jordan Klein, a tech sector specialist at Mizu Securities.

Many bysiders see no reasons to own software no matter how cheap or beaten down the stocks get, Klein wrote in a january fourteenth note to clients. They assume zero catalysts for a re rate exist right now, he said, referring to the potential for higher valuation multiples.

Meanwhile, valuations for software companies keep getting cheaper. The Morgan Stanley basket is priced at eighteen times earnings projected over the next twelve months, its cheapest on record, and well below an average of more than fifty-five times earnings over the past decade. The reason software companies had lofty multiples is because they were subscription based, with recurring revenue that you could extrapolate into the future almost forever, said Osterweiss Capitals Wong.

It's hard to know what multiple they should be trading at if they're going up against AI agents that are running 24-7 and have the ability to complete tasks with big projects getting done in a day. End quote. Did you know that Zipline, the autonomous drone delivery company, didn't start out by delivering packages?

In fact, they actually started out with a robotic toy. We all remember the choices that shaped the course of our lives in business. World-renowned venture capital firm Sequoia Capital calls them crucible moments. Their podcast brings you inside the pivotal decisions that defined some of today's most influential companies. Crucible Moments is entrepreneurial podcasting done right, not just talking endlessly, actual key takeaways you can learn from.

Hosted by Sequoia's Rule of Botha, season three deep dives into the stories behind companies like Zipline, Palo Alto Networks, and Supercell. Crucible Moments is available everywhere you get your podcasts and at Cruciblemoments.com. Go listen to Crucible Moments today. Just because you're a small business doesn't mean you're a small target for bad actors. Cyber criminals know that lean teams often lack the resources to prevent or respond to a breach.

However, even the smallest teams can foil cybercrime with one password. They provide simple security to help small teams manage the number one risk that bad actors exploit. Weak passwords. One Passwords Enterprise Password Manager helps your company eliminate security headaches and improve security by identifying weak and compromised passwords.

And replacing them with strong, unique credentials. Take the first step to better security by securing your team's credentials. Find out more at onepassword.com slash ride and start securing every login. That's onepassword.com slash ride.

According to a court filing, Elon Musk is seeking between seventy nine and one hundred and thirty-four billion dollars in damages from OpenAI and Microsoft, alleging OpenAI defrauded him by abandoning its nonprofit roots and partnering with Microsoft in the first place. Quoting Bloomberg. Musk's lawyer detailed the damages request in a court filing Friday, a day after a federal judge rejected a final bid by OpenAI and Microsoft to avoid a jury trial set for late April in Oakland, California.

Citing calculations by a financial economist expert witnessed C Paul Wazan, the filing says Musk is entitled to a chunk of OpenAI's current five hundred billion dollar valuation after he was defrauded of the thirty eight million dollars in seed money. He donated to OpenAI when he helped found the startup in 2015. OpenAI and Microsoft later disputed the calculations.

Just as an early investor in a startup company may realize gains, many orders of magnitude greater than the investor's initial investment. The wrongful gains that OpenAI and Microsoft have earned, and which mister Musk is now entitled to disgorge, are much larger than Mr. Musk's initial contributions, Musk lawyer Stephen Mollow wrote.

Wazan, the expert witness, calculated the damages request by combining Musk's financial and non monetary contributions, including technical and business advice to OpenAI, according to the filing. He figured that the wrongful gains total sixty-five point five billion dollars to a hundred and nine point four three billion for OpenAI and thirteen point three billion dollars to twenty-five point zero six billion for Microsoft.

Musk's filing says he also plans to seek punitive damages and possibly an injunction that the filing didn't describe. End quote.

The information has an update on the whole Thinking Machines lab stuff, and it continues to be messy. On Wednesday, during an all hands meeting at the AI startup, its CEO, Mira Marathi, announced that she had fired one of Thinking Machines co founders, its chief technology officer, Barrett Zoff, for poor performance and speaking with competitors according to a person familiar with the matter.

During the meeting, two more Thinking Machines researchers, Luke Metz and Sam Schoenholz, dropped bombshells when they separately posted in the company's Slack that they were quitting, the person said. This is Brian cutting in here. We already know that, but we didn't know the details of this happening all at the same time. Back to the quoting.

Marati appeared to be taken aback after the messages from Metz and Schoenholtz, according to the person, while some thinking machine staffers were stunned by the rapid fire disclosures of the three departures. By Thursday, two more thinking machine staffers, researcher Leah Guy and engineer Ian O'Connell, told colleagues that they too were leaving the company. Of the five departures, four have joined OpenAI.

Ordinarily, the departure of about five percent of a startup staff thinking machines employed roughly a hundred people prior to the defections this week wouldn't necessarily be a source of serious concern, but those departures included two of the company's six co-founders. And came after a third co founder, Andrew Tulloch, and the other thing. Left for meta platforms in the fall.

The events rattled investors and thinking machines, according to some of them. It could make it difficult for the one year old startup, which has released only a single product so far, to complete an ambitious funding round that aims to value the company at fifty billion dollars, more than five times its last round. That funding effort is likely to test whether a burst of employee turnover is sufficient to spook private investors in speculative AI startups.

The company's existing investors include Andreessen Horowitz, Assel, and NVIDIA, the AI chip giant. Thinking machines could seek funding from its existing investor base or new ones. One possibility is Google, which is the startup's primary cloud provider. It wasn't so long ago that Miramirati was able to resist the kind of raids that OpenAI performed on thinking machines this week. She had recruited some of her former OpenAI colleagues with an unusual but attractive offer.

thinking machines stock options with strike prices near zero, which they could exercise nearly immediately after joining. Such an arrangement would have been particularly attractive to the founders and early employees. Who joined before investors valued it at ten billion dollars, since their equity would have immediately been worth a lot more within weeks or months.

Zof was one of those early employees. He had joined as Thinking Machine's chief technology officer after a multi-year career at OpenAI during which he ran post training, the crucial last step of model training during which models are ready for release. And Marathi had managed to stave off previous raids on her staff. Over the summer, Meta had offered multiple Thinking Machines researchers large compensation packages. Those researchers declined.

Morati looked like she had enough capital after raising two billion dollars less than a year after starting the company, a nearly unprecedented level for a startup that hadn't released a product yet. Researchers who flocked to the lab in large part came due to Marathi, who was known at OpenAI for her ability to wrangle researchers with big personalities.

Thinking Machines leaders also told them that they'd be able to decide what lines of research they'd pursue and would face less bureaucracy compared to other larger AI labs.

Finally today let me tell you about a big hit AI movie. Quoting the journal. If you could go back in time and document the Manhattan Project, how would you do it? That was the question Demise Hasibas asked his favorite documentary filmmaker nearly a decade ago, long before artificial intelligence was the growth engine of the global economy.

When the leader of Google Deepmind decided to give a movie director extraordinary access to one of the world's leading AI labs, he felt comfortable letting an outsider in on the company's secrets because he knew that Greg Coase had a deeply human measure of success. when he makes movies. I wanna make goosebumps, Coase told me. The thinking game is about both Google Deep Mind and the singular mind of Demise Hasebas.

It's co founder and chief executive who has devoted his life to cracking the mysteries of intelligence. The film tracks his team's revolutionary work, predicting the structure of nearly all known proteins, a breakthrough that was later rewarded with the Nobel Prize in Chemistry. It's both extremely nerdy and very watchable, even if all you know about proteins is that you should be eating more of them.

It also turned out to be a massive hit. The documentary recently went on YouTube and it's already nearing 300 million views. Is that as many as mister Beast? Coase said. In fact, that's more views than any mister Beast's video racked up last year, including the YouTube star's most popular stunt, I spent a hundred hours inside the pyramids.

As it happens, Coase spent hundreds of hours inside another place of wonder. His timing couldn't have been any better. After shooting from twenty eighteen to twenty twenty four, the film went live on YouTube in late twenty twenty five as Google was roaring back to life in the AI wars.

In the past year, one of the world's most valuable companies has nearly doubled in value, and Google's parent alphabet just became the latest tech giant to reach a market capitalization of$4 trillion. If you watch this documentary, you can begin to understand how it got here, end quote. So that's a bit of something for you to do today if you've got today off of work. Usually I do take this holiday Monday off, but they sold ads for the whole week this week, so here we are. Talk to you tomorrow.