Klearcom launched a service that places real phone calls to test AI voice agents, as more companies plan to put them on customer lines.
A company called Klearcom has launched a service that phones your business's AI agent the way a customer would, then reports what went wrong. It is called Agentic Assurance. Klearcom announced it on Nov. 3, 2026. The idea is simple: before customers find the mistakes, an outside tester finds them.
AI voice agents are programs that answer phone calls and talk with callers. Klearcom places real calls over local landline and mobile networks in different countries. It then checks the results against more than 100 measures. These cover sound quality, service quality, compliance and security, and how complex the call is. Testing runs in more than 20 languages.
It also checks each market's rules on telling callers they are talking to an AI, confirming who is calling, and handling personal data. The results become an evidence trail that compliance, risk and audit teams can use.
Klearcom's co-founder and chief technology officer, Satish Barot, says every call to an AI agent can play out differently. He says one update to the AI model or its knowledge base can change its behavior overnight. For that reason, the company says it keeps testing for as long as the agents are talking to customers.
Klearcom argues that no single supplier sees the whole call. The phone carrier, the call center software provider and the AI agent maker each watch only their own part. In its example, a call made within Germany might pass through data centers in Berlin, Paris, Madrid and Frankfurt, while every supplier reports that all is well.
Chief revenue officer Henry Bush says customers blame the brand when an AI agent gets a call wrong. His pitch is that independent evidence from inside each market helps leaders find the fault faster.
Klearcom says 52% of US contact centers plan to roll out AI voice agents in 2027. A contact center is the team and technology that handle customer calls. The company says it tests phone menus, voice bots and AI voice agents across more than 100 countries and 340 carriers.
If your business plans an AI phone agent, ask who will test it from the customer's side, and who will keep testing after every update.
Otto, a new customer-tracking app for field sales teams, lets reps speak their notes instead of typing them. Price and launch details are not yet known.
Read more: Otto launches a sales app you talk to instead of typingWorld models, AI that learns physics from video, are being built into home robots and driverless cars, and your own videos may help train them.
Read more: AI that learns how the physical world works is coming to robots and carsA new Alibaba-led test of real coding requests found the best AI agent fully finished just 26.4 percent of tasks, so human review still matters.
Read more: AI coding helpers fully finish only about 1 in 4 real tasks
