AI Threat Assessment · 26 May 2026

Kore.ai

Enterprise AI Platform
COOKED
6.4/ 10

“Kore.ai spent a decade building the infrastructure for enterprise chatbots, accumulating all the hard-won wisdom about governance, security, and production-ready AI deployment that comes from watching thousands of 'virtual assistant' projects crash into regulatory compliance. The tragedy is that they perfected the art of wrapping guardrails around AI agents just as OpenAI's API made the entire concept of a specialized 'agent platform' about as necessary as a TypeScript wrapper for a calculator.”

Business Model
7.5
Automation Risk
6.0
Moat Strength
4.5
Adaptability
6.5
Need Survival
5.5
AI Threat Level
BUSINESS MODEL REPLACEABILITY

Their Agent Blueprint Language (ABL™) promises 'typed, schema-driven agent behavior' — which sounds impressively engineered until you realise Claude can handle the same enterprise workflows through a ChatGPT Team subscription and a well-written system prompt. The enterprise AI platform that took 10 years to build now competes with 10 lines of OpenAI API calls.

7.5
WORKFORCE AUTOMATION RISK

The 'AI solution architect' roles and enterprise integration specialists are at risk, but someone still needs to convince CISOs that letting Claude loose in the customer service queue won't result in the bot promising refunds for undelivered moon rocks.

6.0
MOAT STRENGTH

Enterprise security, compliance frameworks, and production observability are real moats — banks genuinely cannot just plug ChatGPT into their loan approval process. The problem is that Microsoft, Google, and AWS are building the same governance layer directly into their foundation models, turning Kore.ai's decade of enterprise AI wisdom into a feature checkbox on someone else's platform.

4.5
AI ADAPTABILITY SIGNALS

They launched the 'Artemis' platform with great fanfare about being 'AI-programmable from the ground up' — which reads like a company desperately explaining why their chatbot infrastructure is totally different from everyone else's chatbot infrastructure, honest. The rebrand from 'conversational AI' to 'agentic AI' happened faster than you could update a LinkedIn bio.

6.5
WILL THE NEED SURVIVE AI?

Enterprise AI governance survives — someone needs to make sure the AI doesn't accidentally promise customers a pony with every purchase. But the need for a specialized platform to deliver that governance is evaporating as quickly as foundation model providers add enterprise features that make Kore.ai's core value prop a redundant middleware layer.

5.5
Verdict

“Foundation model providers are absorbing the enterprise governance layer faster than Kore.ai can explain why their 'Agent Blueprint Language' isn't just JSON with aspirations. They built the perfect enterprise AI platform for 2019 and are now watching OpenAI, Anthropic, and Microsoft deliver the same capabilities as API parameters — no PhD in agentic architecture required.”

Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.

Roast another →