Enterprise AI Brief — 2026-09-30

Today: Enterprise AI is moving deeper into production as companies increase spending and deploy autonomous agents, while governance, data foundations, and security become critical constraints.

Top Stories

1. 🤖 OpenAI expands enterprise push with always-on Dots agents

Reuters · September 30, 2026

Bottom line: OpenAI is positioning always-on Dots agents as software that can independently execute multi-step work across business applications.

OpenAI introduced Dots, agents designed to pursue user goals across applications with limited supervision. The agents can work with Slack and Microsoft Teams and draw on OpenAI’s Codex and ChatGPT Work capabilities for research, analysis, document creation, and software development.

Why it matters: The enterprise AI market is shifting from assistants that generate content toward agents that execute workflows. That raises the commercial value of AI while making permissions, auditability, data protection, and human approval controls increasingly important.

🔗 Read the full story

2. 📊 BCG says enterprise AI spending has doubled as value creation accelerates

Boston Consulting Group · September 30, 2026

Bottom line: BCG’s Applied AI Index 2026 finds that corporate AI spending has doubled to 3.3% of revenue, while nearly half of companies now report meaningful AI value.

The research says more than 80% of corporate AI spending now sits outside traditional enterprise IT budgets, indicating that AI investment is spreading into business functions. BCG also projects that agentic AI could account for two-fifths of AI value by 2030.

Why it matters: AI is increasingly being funded as a business transformation capability rather than solely as an IT experiment. The shift also puts pressure on enterprises to connect AI investment with measurable operating outcomes and governance.

🔗 Read the full story

3. 🤖 SAP highlights agentic AI as enterprises move toward autonomous operations

SAP News Center · September 30, 2026

Bottom line: SAP is expanding its enterprise AI strategy around agents that can reason and act across trusted business data and connected processes.

SAP highlighted its Joule Agents, Joule Assistants, Business Data Cloud, and Engagement Cloud as components of an agentic architecture for customer engagement. The company emphasized that security, compliance, governance, and trusted business context are prerequisites as AI takes on more operational responsibility.

Why it matters: Enterprise agent deployment increasingly depends on integration with authoritative business data and process systems rather than standalone LLM capabilities. This reinforces the importance of data governance and controlled agent execution.

🔗 Read the full story

4. 🤖 Salesforce reports rapid growth of agentic search in commerce

Salesforce · September 30, 2026

Bottom line: Salesforce reports that agentic search as the first step in shopping journeys grew 200% year over year, while 52% of Singapore non-adopters plan to deploy agentic AI within six months.

The company’s State of Commerce research combines surveys of 3,450 commerce professionals and 4,690 consumers with behavioral data covering more than 1.5 billion shoppers. Salesforce says only 25% of Singapore organizations report fully unified customer data across sales, service, marketing, and commerce.

Why it matters: AI is becoming a new customer-discovery layer outside traditional search and brand-owned properties. Enterprises therefore need product, customer, and operational data structured well enough for AI systems to represent their businesses accurately.

🔗 Read the full story

5. 🔒 Zscaler reports sharp increase in AI-assisted ransomware data theft

Zscaler · September 30, 2026

Bottom line: Zscaler’s 2026 ThreatLabz Ransomware Report says ransomware-related data theft increased more than 275% year over year to 896.2 terabytes, with attackers increasingly using GenAI and trusted enterprise tools.

The report says blockchain transactions associated with ransomware payments reached $328 million and that manager-level employees and above represented 62% of victims. Attackers are also increasingly abusing tools such as Microsoft Teams and Quick Assist for social engineering, lateral movement, and data theft.

Why it matters: Enterprise AI adoption expands the attack surface while AI also gives attackers faster ways to conduct operations. Security architecture therefore needs to protect both conventional enterprise systems and the new AI-agent and collaboration layers.

🔗 Read the full story

6. 📊 IBM study puts CFOs at the center of enterprise AI transformation

IBM Newsroom · September 30, 2026

Bottom line: IBM’s new global study finds CFOs are taking a larger role in enterprise AI strategy, investment decisions, and converting AI initiatives into operational execution.

IBM’s Institute for Business Value says AI is becoming increasingly integrated into enterprise operations and decision-making, expanding the finance function’s role in determining priorities and capital allocation. The study frames AI adoption as an enterprise transformation issue rather than a technology initiative isolated within IT.

Why it matters: As AI spending moves beyond traditional IT budgets, finance leaders will increasingly need visibility into AI costs, measurable business value, risk, and capital allocation across business units.

🔗 Read the full story