Here are the five biggest tech stories from the last 24 hours: Microsoft's cloud business surges, Alphabet's earnings disappoint despite a doubling cloud unit, SK Hynix posts a record quarter on AI memory demand, GitHub previews security scanning for AI agents, and Cloudflare discloses a cross-customer AI data exposure.
1. Microsoft's Azure Revenue Jumps 48% to $33.2 Billion
Microsoft reported fiscal first-quarter revenue of $92.3 billion, up 19%, beating expectations. Azure revenue rose 48% to $33.2 billion, its first disclosure of Azure revenue in dollars, as cloud and AI demand outpaced new data-center capacity.
Why it matters: The result is the clearest sign yet that Microsoft's massive AI infrastructure spending is converting into sales — and it sets a high bar for rivals reporting in the same earnings wave.
2. Alphabet Shares Fall as Ads Miss, Despite Cloud Doubling
Alphabet reported third-quarter revenue of $127.6 billion, slightly below expectations, as advertising came in about $1.4 billion short. Google Cloud revenue more than doubled to $30.7 billion, but shares fell about 10% after hours on negative free cash flow of $9.1 billion.
Why it matters: Investors are now grading AI spending on cash returns, not just growth: even a doubling cloud business could not offset worries about search, ads, and a rising capital-expenditure bill.
3. SK Hynix Posts Record Quarter as AI Memory Sells Out
SK Hynix reported record third-quarter revenue of 104.6 trillion won and operating profit of 83.3 trillion won, driven by high-bandwidth memory for AI accelerators. The company said it has sold out of HBM and DRAM chips until 2029 and expects shortages to last through 2030.
Why it matters: Memory is now a bottleneck for the whole AI buildout — tight supply lifts chipmakers' profits but could raise costs and delay hardware for everyone downstream.
4. GitHub Previews Scanning for AI-Agent Risks
GitHub previewed code scanning aimed at risks specific to AI-agent applications, checking how autonomous agents use tools, data, and permissions in code. The move targets a new class of vulnerabilities created when software can act on its own.
Why it matters: As agents gain real-world access, a single flawed prompt or permission can do far more damage than a static bug — security tooling has to catch up with software that acts, not just code that runs.
5. Cloudflare Discloses Cross-Customer AI Data Exposure
Cloudflare disclosed a now-fixed isolation failure in its GLM-5.3 Workers AI service between August 27 and September 9, affecting 155 responses with very long prompts. A configuration mismatch and an inference-engine bug let old request data appear in another customer's response, with no evidence of deliberate exploitation.
Why it matters: AI hosting is only as trustworthy as its tenant isolation: customers need proof that one user's prompts cannot leak into another's answers, especially at extreme context lengths.
Sources
- Microsoft Azure revenue surges 48% to $33.2B — GeekWire
- Alphabet shares fall as cash burn worsens — Reuters via KWSN
- SK Hynix profit jumps more than sevenfold — Reuters via Northland News Radio
- GitHub AI security preview — The IT Guys 5 PM recap, October 28, 2026
- Cloudflare cross-customer AI exposure — Precise Impact AI Daily Brief, October 28, 2026