Welcome aboard the 'AI crazy train'
Big Tech is in an AI arms race, each company trying to outspend the others on data centers, GPUs, networking gear, and talent. Engineers can be let go. But the infrastructure? That's permanent. If the AGI dream fades, you're stuck with massive, costly assets.
So when Google announced it would hike capex by $10 billion to $85 billion in 2025 eyebrows went up. Most of it is for things you can't walk back: chips, data centers, and networking.
Google is "jumping aboard the AI crazy train," Bernstein analyst Mark Shmulik wrote, referencing a song by the late bat biter Ozzy Osbourne.
Meta's Mark Zuckerberg brags about Manhattan-sized data centers. And Elon Musk keeps hoarding GPUs. While Sam Altman is building mega-data centers with partners. JPMorgan dubbed this " vibe spending," warning OpenAI might burn $46 billion in four years.
It's no shock when Elon, Zuck, and Sam flex on capex. But Google? That's surprising. "Google doesn't do this," Shmulik said. The company has been viewed as measured in recent years, prioritizing investment intensity with care. Not anymore.
Now investors want to know: Will these swelling bets pay off?
There are promising signs. Since May, Google's monthly token processing (the currency of generative AI) has doubled from 480 trillion to nearly a quadrillion. Search grew 12% in Q2, beating forecasts. Cloud sales surged 32%. CEO Sundar Pichai said Google is ramping up capex to support all this growth.
But it's still a huge gamble. "Does the current return on invested capital seen in both Search and Cloud hold up at higher [capex] intensity levels," Shmulik asked, "or is the spend a very expensive piece of gum trying to plug an AI-sized hole?" He leans optimistic.
Still, Google shares rose just 1% after these results. Not exactly a resounding endorsement.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
26 minutes ago
- Forbes
New Models From OpenAI, Anthropic, Google – All At The Same Time
It's Christmas in August – at least, for those tech-wonks who are interested in new model releases. Today's news is a very full stocking of brand new LLM editions from three of the biggies – OpenAI, Anthropic, and Google. I'll go over these one by one, discussing what these most recent model iterations bring to the table. OpenAI OSS Models First, the tech community is getting an eye on OpenAI OSS 120b and OSS 20b, the first open-weight systems from this company since ChatGPT 2. Now, coverage from Computerworld and elsewhere points out that, although these models have Apache licenses, they are not fully open source in the conventional way, ,but partly open: the weights are open source, while the training data is not. Powered by one 80GB GPU chip, the larger OSS models, according to the above report, 'achieves parity' with the o4-mini model vis a vis reasoning power. The smaller one can run on smartphones and other edge devices. The models come quantized with MXFP4, a low-precision data type for accelerating matrix multiplications. Let Them Work Another interesting aspect of the new OSS models has to do with chain of thought, something that has revolutionized inference, while raising questions about comparative methodology. Basically, we want the LLMs to be accurate, but engineers have found that, in many cases, restricting or overly guiding systems causes them to 'hide' CoT. So OpenAI has chosen not to optimize the models in this way. 'OpenAI is intentionally leaving Chain of Thought (CoTs) unfiltered during training to preserve their usefulness for monitoring, based on the concern that optimization could cause models to hide their real reasoning,' writes Roger Montti at Search Engine Journal. 'This, however, could result in hallucinations.' Montti cites the following model card report from OpenAI: 'In our recent research, we found that monitoring a reasoning model's chain of thought can be helpful for detecting misbehavior. We further found that models could learn to hide their thinking while still misbehaving if their CoTs were directly pressured against having 'bad thoughts.'…In accord with these concerns, we decided not to put any direct optimization pressure on the CoT for either of our two open-weight models. We hope that this gives developers the opportunity to implement CoT monitoring systems in their projects and enables the research community to further study CoT monitorability.' So, the models are allowed to have these 'bad thoughts' in aid of, I suppose, transparency. OpenAI is then honest about the higher chance for hallucinations, so that users know that this trade-off has been made. Claude Opus 4.1 Here's how spokespersons rolled out the announcement of this new model Aug. 5: 'Today we're releasing Claude Opus 4.1, an upgrade to Claude Opus 4 on agentic tasks, real-world coding, and reasoning. We plan to release substantially larger improvements to our models in the coming weeks. Opus 4.1 is now available to paid Claude users and in Claude Code. It's also on our API, Amazon Bedrock, and Google Cloud's Vertex AI. Pricing is the same as Opus 4.' What's under the hood? The new Opus 4.1 model ups SWE-Bench Verified marks, and boosts agentic research skills. A breakdown of capabilities shows a 2-point increase in SWE-based agentic coding (72.5% - 74.5%) and improvement in graduate-level reasoning with GPQA Diamond (79.6% - 80.9%) over Opus 4, and slight increases in visual reasoning and agentic tool use. For a model set that pioneered human-like user capabilities, this continues to push the envelope. As for strategy: 'The release comes as Anthropic has achieved spectacular growth, with annual recurring revenue jumping five-fold from $1 billion to $5 billion in just seven months, according to industry data,' writes Michael Nunez at VentureBeat. 'However, the company's meteoric rise has created a dangerous dependency: nearly half of its $3.1 billion in API revenue stems from just two customers — coding assistant Cursor and Microsoft's GitHub Copilot — generating $1.4 billion combined. … The upgrade represents Anthropic's latest move to fortify its position before OpenAI launches GPT-5, expected to challenge Claude's coding supremacy. Some industry watchers questioned whether the timing suggests urgency rather than readiness.' Regardless, this is big news in and of itself, including for the millions of users who rely on Claude for business process engineering or anything else. Genie 3 This is the latest in the series of Genie models coming out of Google's DeepMind lab that create controlled environments. In other words, this is a gaming world model. Proponents of the new model cite longer-term memory over Genie 2's limit of about 10 seconds, as well as better visual fidelity and real-time responses. 'DeepMind claims that the new system can generate entire worlds that you can interact with consistently for several minutes in up to 720p resolution,' reports Joshua Hawkins at BGR. 'Additionally, the company says that the system will be able to respond to what it calls 'promptable world events' with real-time latency. Based on what the videos show off, it seems like Google has taken a major step forward in creating entire video game worlds using AI.' 'Genie 3 is the first real-time interactive general-purpose world model,' said DeepMind's Shlomi Fruchter in a press statement according to a TechCrunch piece suggesting that the lab considers Genie 3 to be a 'stepping stone to AGI,' a big claim in these interesting times. 'It goes beyond narrow world models that existed before. It's not specific to any particular environment. It can generate both photo-realistic and imaginary worlds, and everything in between.' All of these new models are getting their first rafts of public users today! It's enough to make your head spin, especially if you're responsible for any kind of implementation. What do you choose? To be fair, there is some amount of specialization involved. But many professionals closest to the industry would tell you it's the speed of innovation that's challenging: given the track record of most companies, by the time you get something worked into business operations, it's likely to already be obsolete! Stay tuned.
Axios
an hour ago
- Axios
Truth Social's Perplexity search comes with Trump-friendly media sources
President Trump's social media company Truth Social unveiled a new search tool powered by AI answer engine Perplexity on Wednesday — but Truth Social users who run Perplexity searches may find their results limited to a narrow set of typically Trump-supporting media outlets. Why it matters: Increasingly, where you ask online matters as much as what you ask. Catch up quick: Trump Media & Technology Group on Wednesday said it was launching a public beta test of a search engine, Truth Search AI, powered by Perplexity. Perplexity has been seen as a nascent Google-killer and is often touted by investors as a possible acquisition target for the likes of Apple. How it works: Axios asked seven questions on both a logged-in Truth Social account and the free, logged-out Perplexity website … What happened on January 6, 2021? Why was Donald Trump impeached? What crimes was President Trump convicted of? Did Donald Trump lose the 2020 election? What is Hunter Biden's laptop a reference to? Was Hillary Clinton ever charged with a crime? Is the new "Naked Gun" movie good? Between the lines: In most cases, the responses were generally similar — but the sources linked to the answers were not. In all seven responses on Truth Social, either was the most common, or the only, listed source of information. Other sources were Washington Times or Epoch Times. In contrast, answers via the public version of Perplexity returned a wider variety of sources, including Wikipedia, Reddit, YouTube, NPR, Esquire and Politico. Although the questions were matched and asked at roughly the same time, there was no source overlap. What they're saying: A Perplexity spokesperson tells Axios that Truth Social is a customer of Perplexity's API, which means it — like tens of thousands of other developers — is building tools to its own specifications, and with its own restrictions. Any customization, like limiting the sources for its answers, would happen entirely on the Truth Social side. While it's standard practice for platforms to put their own layers of rules and information on top of tools, search tools usually cast a broader net. Truth Social did not mention any restrictions in its announcement, although it did say it plans to "refine and expand our search function based on user feedback." Perplexity's Sonar API specifically includes the ability for users to customize sources, which the company noted in January was a top user request. The bottom line: When you ask a search tool a question, particularly in the age of AI, it's best to know exactly where your information is coming from, and whether there are any limits on what the tool will tell you. Expect more of this as governments and businesses increasingly put their thumbs on the AI scale to serve their interests.
Yahoo
an hour ago
- Yahoo
OpenAI's $500 Billion Power Play: Is This the Hottest Deal in Tech Right Now?
OpenAI is weighing a massive secondary stock sale that could push its valuation to around $500 billion, according to people familiar with the matter. The deal still in early discussions would allow current and former employees to cash out some of their shares, with existing investors like Thrive Capital reportedly circling. If completed, this would mark a roughly 67% jump from OpenAI's previous $300 billion valuation tied to a $40 billion raise, one of the largest in private tech history. The move isn't just about rewarding staff it's also a strategic play to retain top talent in the face of fierce poaching from rivals like Meta, which has dangled nine-figure packages to lure researchers away. Warning! GuruFocus has detected 7 Warning Sign with MSFT. The timing is no coincidence. OpenAI recently secured $8.3 billion in a second tranche of that same $40 billion round, which one source said was oversubscribed by about five times. This fresh capital comes as the company doubles down on product and platform expansion. ChatGPT usage is climbing fast reportedly hitting 700 million weekly active users while the app now processes over 3 billion messages per day. OpenAI is also preparing its next major release, GPT-5, which is undergoing internal testing. While no official launch date has been announced, the model is expected to reinforce OpenAI's lead amid growing competition. On the hardware front, the company is moving ahead with a nearly $6.5 billion all-stock acquisition of a device startup co-founded by Apple design veteran Jony Ive. But the company isn't without friction points. Key investors are still in discussions over how OpenAI should be structured going forward including Microsoft (NASDAQ:MSFT), which has poured in more than $13 billion. At the heart of those talks is Microsoft's stake and long-term access to OpenAI's tech, with the current deal set to run through 2030. As OpenAI races to define its next chapter spanning AI platforms, consumer hardware, and employee retention investors are left with a familiar question: how much more upside is still on the table? This article first appeared on GuruFocus. Melden Sie sich an, um Ihr Portfolio aufzurufen.
