Agents in the Enterprise: Salesforce and DeepL See Productivity Move the Needle as Use Cases Grow for AI

Why It Matters

Enterprise AI agents promise measurable productivity gains and lower operational risk, making them a strategic differentiator for companies that can deploy and monitor them at scale.

Key Takeaways

• 70% of Heathrow interactions now handled by agents
• DeepL reports 2,000 enterprises deploying AI agents
• Agents can be deployed within weeks, not months
• Contextual intelligence reduces guesswork, speeds decisions
• Observability tools ensure transparent, auditable agent actions

Pulse Analysis

The shift from experimental pilots to production‑grade AI agents marks a pivotal moment for enterprise technology. In 2025, firms tested agentic AI in isolated use cases; by 2026, vendors like Salesforce and DeepL are demonstrating that agents can be integrated across entire business units within days or weeks. This accelerated timeline is driven by improved model reliability, pre‑built data pipelines, and a focus on tangible productivity metrics, positioning AI agents as a mainstream tool rather than a niche experiment.

Salesforce’s rollout at Heathrow illustrates how agents can replace large swaths of human interaction while delivering 24/7 service. The company’s Agent Script language couples deterministic business rules with generative reasoning, while Data 360 injects real‑time contextual data, allowing agents to draw accurate conclusions quickly. DeepL’s global customer base of 2,000 enterprises underscores the breadth of applications—from automating report synthesis to extracting key contract terms—highlighting the versatility of agents in freeing staff from repetitive tasks.

Observability has emerged as the linchpin for enterprise confidence in AI agents. Solutions like MuleSoft’s Agentforce Observability give operators granular insight into each decision, addressing regulatory and risk concerns that have hampered broader adoption. Financial institutions such as Visa are already mandating transparent AI workflows, signaling that trust and auditability will be as critical as raw performance. As agents become embedded infrastructure, they will recede into the background, delivering continuous value without drawing attention, much like the digital utilities that power modern enterprises.