Full Summary
This Saturday, August 29th, the "agent economy" is scaling faster than it can be metered, creating significant cost governance challenges for businesses, according to IDC. A staggering 95% of enterprises globally now use at least one company-funded agent-enabled workflow, with the average organization running about 11. Both IDC and Tech Critter highlight that enterprises with cost visibility report an average monthly spend of over $117,000 on agent inference and related services. The Motley Fool adds that the number of active agents is projected to reach 2.5 billion by 2030, a nearly 80-fold increase from 2025. However, 67% of enterprises ran over their agent spend budget by more than 10% in the past year, with 24% significantly over forecast, IDC reports. BankInfoSecurity warns that 56% of organizations already report exposure to agentic AI, with 91% unable to stop a risky agent action before it happens. Plus, 65% admit their current zero-trust controls don't cover non-human identities. On the development front, Intel has unveiled three new hardware architectures designed for agentic AI workloads, TechRadar confirms. These include the 256-core Diamond Rapids Xeon, set for 2027, and the Crescent Island inference GPU. Gnani AI has also launched a new large language model, Evon v3.3, and an agentic AI platform called Plexus, APAC Media states. Meanwhile, AI coding agents are evolving beyond simple autocomplete, now able to examine repositories, modify files, run tests, and correct errors to achieve a desired goal, according to Towards Data Science. HackerNoon demonstrates that production-ready AI agents can be built using existing web technologies like PHP, MySQL, and cPanel, challenging assumptions about complex infrastructure. This rapid expansion means your data and company resources are increasingly managed by autonomous AI, requiring businesses to adapt their security and financial oversight to prevent costly overruns and potential breaches.