From SiliconANGLE · Zeus Kerravala · · 2m
NiCE Cognigy focuses on running AI agents, not building them
NiCE's Cognigy platform makes AI agents dependable without complex model building.
In brief
NiCE's Cognigy platform makes AI agents dependable without complex model building. NiCE's next-generation Cognigy platform focuses on reliably running AI agents across customer journeys rather than building them, with 16-hour autonomous operation at 50% success rate. Originally reported by SiliconANGLE.
The AI agent debate
The contact center has been the low-hanging fruit for AI, but the industry has spent two years arguing over which AI agent is the smartest.
“The contact center has been the low-hanging fruit for the artificial intelligence industry”
Dependability over intelligence
NiCE's next-generation Cognigy platform focuses on making AI agents dependable rather than building them, with 16-hour autonomous operation at 50% success rate.
“Capable and dependable do not always mean the same thing”
Simplicity as strategic fix
Cognigy lets builders stand up agents in a few clicks with pre-provisioned models, targeting smaller businesses and contact center teams.
“The large number of options when configuring an agent creates friction to get started”
Hybrid deployment approach
Most enterprises use hybrid deployment where agents handle some tasks autonomously while others remain deterministic.
“Hybrid is actually where most of our customers are”
Observability features
New tracing shows each turn of an agent, including which skills loaded and tools called, with multivariate testing for up to 10 variants.
“You can have the best builder experience but if you don't have any feedback loop on what you're building you're never going to build a great agent”
Voice-to-voice support
Cognigy supports voice-to-voice conversations that sound more natural than text-to-speech cascades.
“Voice-to-voice models sound just better”
Agent identity and security
Cognigy acts as the front door for personal agents, requiring authentication before they can proceed with customer interactions.
“Cognigy is the front door of the contact center where a consumer's personal agent must authenticate before it can proceed”
Industry collaboration
NiCE is co-developing the Personal Agent Protocol with Meta and Sierra, a direct competitor.
“NiCE is co-developing the Personal Agent Protocol with Meta and Sierra”
Governance challenges
Lowering the barrier to build agents means more governance features are needed to manage the growing number of agents.
“The more you add the more bloated your agent becomes”
Model selection strategy
Boutier recommends comparing variants by cost per resolved conversation rather than anecdotal preferences.
“Compare variants by cost per resolved conversation not by anecdote”
Compliance requirements
Enterprises must build in regional ai disclosure and human escalation rules to meet compliance standards.
“Regional rules on ai disclosure and human escalation must be built in by the customer”
Future of personal agents
NiCE advises accepting personal agents early to avoid being left behind as they become mainstream.
“Don't ban personal agents rather set policies now on which personal agents you'll accept”
The Cognigy strategy
NiCE bets on dependability and governance as the key value for ai agents rather than raw intelligence.
“Intelligence is getting cheaper dependability governance and the ability to blend autonomous reasoning with deterministic control are not”
Market position
NiCE aims to make Cognigy the core component of its platform as personal agents become mainstream.
“The Cognigy acquisition will move from an add-on to a core component of the NiCE platform”
The bottom line
NiCE's next-generation Cognigy platform focuses on reliably running AI agents across customer journeys rather than building them, with 16-hour autonomous operation at 50% success rate.
