Business Technology News Roundup: Apr 06, 2026

On March 31, OpenAI officially closed the largest private funding round in the history of the technology industry, $122 billion, at a post-money valuation of $852 billion. Amazon put in $50 billion, Nvidia contributed $30 billion, and SoftBank added $30 billion as its second tranche of a previously announced commitment. The round also included participation from retail investors for the first time.

To put the company's growth in perspective: OpenAI crossed $25 billion in annualized revenue at the end of February 2026, up from $21.4 billion at year-end 2025 and roughly $6 billion at the end of 2024. That kind of growth, a roughly 4x jump in just 14 months, is unlike anything seen in enterprise software history. ChatGPT now has over 900 million weekly active users and more than 9 million paying business customers.

The big news going forward is what comes next. OpenAI is preparing to go public as early as Q4 2026, and is reportedly targeting a listing valuation of up to $1 trillion, which would be the largest public offering in history, surpassing Saudi Aramco and Alibaba. In a surprising move, the company confirmed it is discontinuing Sora, its standalone AI video-generation app, after user numbers reportedly halved from one million to under 500,000. Additionally, the company is shelving experimental "side quests", including a controversial adult-themed chatbot, to focus on its core enterprise business and the development of a "unified AI superapp."

There's a catch though. Despite all that revenue, OpenAI is not profitable and doesn't expect to reach breakeven until 2030, with annual cash burn projected to reach $57 billion by 2027. So what you're really buying into, if this IPO happens, is a massive bet that enterprise AI adoption keeps scaling fast enough to close that gap. Whether public market investors have the appetite for that story is the real question.

On the morning of March 31, thousands of Oracle employees across the US, India, Canada, and Mexico woke up to termination emails from "Oracle Leadership." The emails arrived at approximately 6 a.m. local time, with no prior warning from HR or direct managers. Access to company systems was cut immediately.

Investment bank TD Cowen estimates the cuts could total between 20,000 and 30,000 workers, roughly 18% of Oracle's global workforce of about 162,000. Among the hardest hit units were Oracle's Revenue and Health Sciences division and its SaaS and Virtual Operations Services group, each reportedly seeing cuts of at least 30%.

Here's the part that's hard to ignore: Oracle posted a 95% jump in net income last quarter, reaching $6.13 billion. This is not a company in trouble. It's a company making an aggressive, expensive bet on AI infrastructure and funding that bet by eliminating a huge portion of its workforce. Oracle has committed to an aggressive AI infrastructure buildout that requires an estimated $156 billion in capital spending, according to TD Cowen, and the company has been leaning on the debt market to fund it.

The broader story here isn't just Oracle. Amazon, Meta, Atlassian, and Microsoft have all made significant layoffs in recent months, with the same explanation attached: redirecting resources toward AI. The displacement happening in the tech industry right now isn't theoretical anymore. It's showing up in people's inboxes on Tuesday mornings.

On April 2–3, Google launched Gemma 4, its most capable open-source AI model family to date. Built from the same research and technology as Gemini 3, Gemma 4 is released in four sizes: Effective 2B, Effective 4B, 26B MoE, and 31B Dense. And is available under a commercially permissive Apache 2.0 license.

What makes this a big deal is the combination of capability and accessibility. The 31B model currently ranks as one of the top open models in the world on standard benchmarks, outcompeting models 20 times its size. And because it's genuinely open-source under Apache 2.0, anyone from individual developers to large enterprises, can download it, modify it, run it on their own hardware, and deploy it commercially without paying Google anything.

Gemma 4 can run locally on billions of Android devices and select laptop GPUs, and has been downloaded from platforms like Hugging Face, Kaggle, and Ollama. The model handles text, images, video, and audio, supports over 140 languages, and is designed for agentic workflows, meaning it can handle multi-step tasks and interact with external tools and APIs on its own.

For businesses that have been hesitant about sending sensitive data to third-party AI services, this is worth paying attention to. Running a capable model entirely on your own infrastructure, with no subscriptions and no data leaving your systems, is now a realistic option.

This week also brought into sharper focus just how much the Trump administration's sweeping tariff policy stands to disrupt the tech industry, and how quickly companies are scrambling to react.

Nearly all major U.S. tech companies manufacture products in China, including Apple, Microsoft, Amazon, Google, HP, Dell, Intel, Cisco, and Tesla. Rising tariffs on Chinese imports are expected to drive up prices on smartphones, laptops, tablets, and consumer electronics across the board.

Apple is arguably the most exposed. At least half of its iPhones are made in China, and Apple has begun making plans to shift all iPhone production to India by the end of 2026. But that move takes years to execute and costs billions. Apple has increased its total U.S. investment commitment to $600 billion as part of an American Manufacturing Program, an assertive attempt to align with Washington's industrial policy and reduce tariff exposure.

For consumers, the message is simple: things are going to cost more. Analysts have estimated that an American-made iPhone could retail between $1,500 and $3,500, depending on how much of the supply chain gets reshored. Even if that extreme scenario doesn't happen, the combination of tariffs on imported components, rising memory chip prices, and ongoing supply chain restructuring all push in the same direction, higher prices on almost every piece of tech hardware you buy. This is one to watch closely as the year progresses.

A sophisticated cyberattack targeted the Patriot Regional Emergency Communications Center in northern Massachusetts this week, disrupting public safety and emergency computer systems for several municipalities. The breach, which began on Tuesday, forced local government officials to take critical systems offline and issued an urgent warning for users to change passwords on the CodeRED emergency alert platform. While voice-based 911 services remained operational, the administrative and dispatch backends were severely throttled, highlighting vulnerabilities in regional infrastructure.

This incident serves as a grim reminder that "soft targets" in local government remain the Achilles' heel of US national security. While the federal government focuses on protecting the power grid and financial systems, small regional hubs often lack the budget for the advanced defensive tools necessary to stop modern ransomware groups. The event is already being cited in D.C. as a reason to accelerate the Cyber Strategy for America, which calls for closer public-private coordination and streamlined regulations to protect critical local infrastructure from transnational cybercrime networks.