Microsoft Loses $360B as AI Bills Scare the Wall Street Microsoft stunned Wall Street as its stock fell about 10% to 12% , erasing roughly $360B to $400B in value intraday even after it posted a revenue beat. Microsoft's capital expenditures hit $37.5B, up 66%, as it races to build AI data centers and buy scarce chips. While the quarter's performance is not being punished, investors are concerned about the company's high capital expenditures, fearing that the company may not see a return on its investment in AI technology in the near future. Simple read on what spooked the market: Contrast: Meta rose 10.4%, while Oracle fell 2.2%, showing investors did not punish “AI” evenly.
The market is no longer willing to overlook promises of future success in exchange for high valuations. Companies like Microsoft will need to demonstrate concrete evidence of success in AI monetization in order to regain investor confidence and avoid further punishment in the stock market. The next earnings season will be crucial in determining whether tech giants can deliver on their promises and prove that their investments in cloud and AI technologies are paying off. Google Puts Gemini 3 ‘Auto Browse’ in Chrome for Pro Users Google began a US preview of Gemini 3 “ Auto Browse ” inside Chrome’s side panel.
The feature can click, scroll, open tabs and type to finish tasks like shopping comparisons and form filling. Google says it built this with security and control by design, but if it misreads a page or gets steered by a shady popup, a wrong click can start being a cost. Google assures that it has implemented security measures, it is important for users to remain vigilant while using this new browsing tool. Here is what the rollout includes: Price: Auto Browse sits behind Google AI Pro at $19.99 a month or Google AI Ultra at $249.99 a month.
How it runs: It works from a Gemini side panel and uses Google Password Manager only if you allow it.
Chrome is chasing the same reward as OpenAI and Perplexity: fewer clicks, more outcomes. If Google keeps the agent on a short leash, this becomes a real quality of life upgrade for busy work. If it normalizes autonomous browsing, scammers and dark patterns get a new target. In the end, the success of this new browsing agent will depend on how well Google is able to monitor and control its actions.
OpenAI Seeks $100B From Amazon, Microsoft and NVIDIA OpenAI is trying to raise a massive new round with three familiar names showing up. Reports say NVIDIA, Microsoft and Amazon are discussing investments that could total up to $60B and OpenAI is said to be close to getting term sheets. These checks may come with business strings, like where OpenAI rents servers and who gets to resell its products. If OpenAI secures additional funding, it will be able to continue training at full speed.
Here is what the reporting points to: Split: NVIDIA up to $30B, Microsoft under $10B, Amazon more than $10B and possibly over $20B. Terms: Amazon's terms involve increased cloud spending and the sale of enterprise ChatGPT. Power: OpenAI and NVIDIA already talk about 10 gigawatts of AI data centers and up to $100B tied to rollout.
This deal resembles a common cycle in the AI industry: investors fund the model, which then acquires more chips and cloud resources, all while being seen as progress by all parties involved. Increased funding can lead to quicker product launches and more affordable prices for users. It can also get chaotic because when your biggest backers also sell the meter you run on, independence turns into a pricing negotiation. Success in this context depends on the entity possessing the greatest influence over distribution and infrastructure.
It consolidates messages, comments, and reviews into a unified workflow, integrating reporting and social listening to transform everyday conversations into actionable insights. Core functions (and how to use them): Smart Inbox triage: Pull DMs and comments into one queue, assign threads to teammates, and close conversations so nothing gets double answered or forgotten. Message tagging: Create tags like “pricing,” “bug,” and “feature request,” then tag incoming messages to build a weekly list of what people actually ask for. Publishing calendar: Draft posts, schedule them across channels, and keep approvals in one place so your posting plan is visible and repeatable.
Reporting exports: Generate a simple weekly report showing top posts, growth, and engagement so you can prove what worked and stop guessing. Social listening basics: Track keywords for your brand, product names, and competitors to catch sudden spikes, recurring complaints, and emerging topics worth a post. Try this yourself: Set up a 10-minute “content from inbox” loop. Establish 5 tags: Pricing, Bug, Feature, Confused, and Love.
Tag the following 20 messages before responding. Afterwards, export or record the tag frequencies. Develop a post concept based on the most frequently used tag, and utilize the top response as your initial draft.