AI observability stories - Page 2

UK CIOs embed AI agents in critical workflows but lack real-time oversight as board pressure mounts for provable returns and tighter audits.

Kyndryl launches policy-as-code guardrails to keep agentic AI workflows governed, auditable and compliant in tightly regulated sectors.

New research warns that nearly half of over 3 million AI agents in US and UK firms run ungoverned, exposing data and systems to rising risks.

Forrester warns firms chasing AI hype risk failure and rising costs if weak IT foundations, governance and data are left unresolved.

Snowflake launches Semantic View Autopilot and wider AI tooling to speed enterprise AI from experimentation into governed production use.

Mphasis wins AWS AI Competency, bolstering its generative AI credentials and signalling readiness to scale enterprise deployments on AWS.

Dynatrace launches Intelligence platform with agentic AI, deeper cloud hooks and next-gen RUM to tackle escalating data and ops complexity.

Employees trust AI data more than ever, yet European and UK firms admit literacy gaps and weak governance are undermining safe adoption.

Security and governance fears keep half of agentic AI stuck in pilots, even as leaders raise budgets and pursue tightly supervised use cases.

Teradata launches Enterprise AgentStack, a unified toolkit to help large organisations scale governed AI agents across hybrid estates.

New Relic data links use of its AI observability tools to 27% less alert noise, 25% faster fixes and up to fivefold higher deployment rates.

AI agents surge into big business with scant oversight, leaving governance, security and trust frameworks struggling to keep pace.

New Relic debuts observability tools for ChatGPT-embedded apps, tackling i-frame blind spots with telemetry, session replay and UX analytics.

e& enterprise and Emergence AI team up to roll out autonomous agentic AI for regulated organisations across MENAT and Türkiye.

F5 rolls out AI Guardrails and AI Red Team to harden runtime security, blending adversarial testing with real-time policy enforcement.

Agentic AI promises rapid software gains, but a recent Claude Code cyberattack shows how “double agents” can outpace unprepared developers.

Nintex warns enterprise agentic AI will face a 2026 reality check, with tighter budgets, targeted use cases and tougher governance.

By 2026, Java is tipped to be AI's production backbone, driving heavier compute, tighter security and runtime modernisation.

Pendo rolls out Agent Analytics to help companies track AI agents in real workflows, spot friction points and turn usage data into improvements.

Tetrate has unveiled Agent Router Enterprise, a platform to certify AI agents as production ready with shared standards, guardrails and metrics.