Tech companies are finally generating real revenue from their AI investments, but they're pouring that money — and more — back into data centers in a bet that the AI boom is just getting started.

Why it matters: Investors will want to see those massive investments not only generate revenue, but profits too.

• While investors applauded strong returns this quarter from Google, Microsoft and Meta, there are signs that their patience is not infinite.
• Light guidance from Amazon sent shares plunging yesterday after hours, with investors grilling CEO Andy Jassy over its AI strategy and the growth rate of the firm's cloud service — and apparently they were not thrilled by his answers.

Driving the news: Microsoft and Google credited the AI boom with helping them deliver better-than-expected results for the most recent quarter.

• Investors weren't wowed by Amazon's outlook, but the Web Services business generated nearly $31 billion in quarterly revenue, about what analysts were expecting.
• Shares of Meta surged after CEO Mark Zuckerberg unveiled his personal superintelligence strategy, the thesis underlying the massive investments that he's making not only in computing horsepower, but also in talent.
• Apple delivered a strong quarter, though that was thanks to iPhone sales. The company has struggled to deliver on Apple Intelligence, with some features promised last year having yet to ship and only modest advances included in the latest versions of the software that powers its devices.
• OpenAI — still a private company and not subject to quarterly disclosures — is also seeing continued, rapid growth.

Between the lines: The tech giants are all spending huge on data centers and computing horsepower, helping Nvidia become the first company with a $4 trillion market capitalization, and Microsoft the second.

• OpenAI is expanding its Stargate data center network, announcing plans yesterday for the first European project: Stargate Norway.
• Meta, which has been making headlines for its eye-popping salary offers to top AI engineers, is also spending big on computing capacity.
• Microsoft expects to spend $100 billion in capital expenses next year, including a record $30 billion this quarter, mostly for AI.
• All told, big tech companies invested roughly $245 billion in AI infrastructure last year and are seen upping that to $320 billion this year, per numbers compiled by the Financial Times.

What they're saying: "Meta's blowout quarter was the best example to date of AI having a tangible impact on revenue and earnings growth at scale," longtime tech analyst Gene Munster wrote this week. "The question of whether AI has an ROI can be put to rest."

The bottom line: AI is fueling revenue, but spending is growing even faster.

Sensitive corporate data appeared in more than 4% of generative AI prompts and over 20% of uploaded files in the second quarter of this year, according to new research from Harmonic Security released yesterday.

The big picture: The problem isn't new, but as workplace genAI use increases, many employers still lack or don't enforce AI policies, causing employees to use bots in secret or without proper training.

By the numbers: Harmonic Security sampled a million prompts and 20,000 files submitted to 300 genAI tools and AI-enabled SaaS applications between April and June.

• 43,700 of the prompts (4.4%) and 4,400 of the uploaded files (22%) contained sensitive information.

Between the lines: Personal and free chatbot accounts make up a large share of corporate data exposure.

• Nearly half (47.4%) of sensitive uploads to Perplexity were from users with standard (non-enterprise) accounts.
• About a quarter of the prompts with sensitive information came through the free version of ChatGPT, and another 15% of sensitive prompts were submitted via free versions of Google Gemini accounts.

Zoom in: Overall, including free and paid tiers, ChatGPT was by far the biggest source of prompt-based information exposure, followed by Microsoft Copilot and Google Gemini.

Code was the most common type of sensitive data sent to chatbots.

• Harmonic says code was "especially prevalent in ChatGPT, Claude, DeepSeek and Baidu Chat."
• The number of prompts containing proprietary code was disproportionately high in Claude, which is often regarded as the best AI tool for coders.
• Sensitive prompts to ChatGPT involved M&A planning, financial modeling, and investor communications.

The intrigue: Tools that feel safe — like document editors or design platforms — may now include genAI features trained on user data, creating exposure risk that bypasses traditional controls, Harmonic says.

• Harmonic found that Canva, Replit, Grammarly, and other tools with LLMs embedded inside them were used for legal strategy, internal emails, client data, and code.
• These uses were often not flagged as AI tools by corporate systems.

The fine print: The organizations Harmonic looked at have all deployed the company's tools to secure their data, meaning actual exposure elsewhere could be even higher.

Apple's better-than-expected quarterly revenue and earnings come amid strong iPhone sales, driven in part by people pulling forward purchases to avoid tariffs.

Why it matters: Apple is seen as a bellwether for the tech industry, and its results are closely tied to the fortunes of a host of other companies, including component suppliers and wireless carriers.

Driving the news: Apple earned $23.4 billion, or $1.57 per share, on revenue of $94.0 billion for the three months ended June 28.

• Analysts were looking for per-share earnings of around $1.43, on revenue of $89.22 billion, per Bloomberg estimates.

The intrigue: CEO Tim Cook said on an earnings call that about a tenth of the company's revenue growth could be attributed to customers pulling forward purchases due to tariff fears, a phenomenon seen across industries this past spring.

• Tariffs cost the company about $800 million in the June quarter, and are projected to cost about $1.1 billion this quarter.
• President Trump has focused some ire on Apple over plans to build iPhones in India, as opposed to moving production to the U.S.

Keep reading.

• OpenAI has removed a feature that allowed some shared ChatGPT conversations to be made public after some highly personal queries made their way onto Google. (OpenAI on X)
• Figma shares more than doubled after the company's IPO. (Axios)
• Gemini 2.5 Deep Think — the AI model that achieved gold medal performance at the International Mathematical Olympiad — is now available to all Google AI Ultra subscribers. (TechCrunch)
• Former Meta civil rights chief Roy Austin Jr. is joining Howard University as the inaugural director of the Howard Law Artificial Intelligence Initiative.

It's been a big month for sea mammal rescues in the Bay Area. On July 19, volunteers and officials helped corral a young sea lion that was on a San Rafael road, while earlier in the month another sea lion was found wandering around a Petaluma outlet mall.