Skip to content

700,000 calls, 15 reviewers a day: xAIa's case for ComplAI

One xAIa client generated 700,000 recorded calls in a month while its compliance team could clear 15 to 20 a day. The Dubai company built ComplAI, which it says reviews every conversation and cuts manual review effort by more than 80%.

By 4 min read
0

A month of calls, a team of reviewers

One deployment of xAIa's technology, built for a UK energy brokerage, generated 700,000 recorded conversations in a single month, while the brokerage's human compliance function could review at most 15 to 20 calls a day, according to a 25 July 2026 blog post from xAIa, a Dubai company that builds AI products for enterprises across the Middle East and North Africa.

The brokerage was closing roughly 120 regulated deals a day over the period the post describes, with as many as 50,000 calls handled between 9am and 6pm on its busiest days, xAIa says. Every one of those calls, along with the chats and documents tied to each deal, fell under compliance rules that had to be checked before a sale could stand.

Run the reviewer capacity against the call volume and the mismatch is stark. Clearing 700,000 calls at a rate of 15 to 20 a day would take a single reviewer roughly 35,000 to 46,667 working days on its own. Spread across a typical 22-day working month, closing that gap with people alone would mean staffing somewhere between about 1,600 and 2,100 full-time reviewers, not the handful most compliance teams actually run. That is the arithmetic xAIa says drove the decision to build an automated layer instead of trying to hire around the problem.

There is no headcount that closes that gap, at any budget, in any market. — Team xAIa, xAIa company blog

What ComplAI checks

xAIa's response was a product it calls ComplAI. On ComplAI's product page, the company says it reads every recorded call, chat and document against a client's compliance rules and returns a verdict with the supporting evidence attached, rather than leaving a reviewer to search for it themselves.

For the energy brokerage, xAIa built an initial ruleset covering eight standards and twenty-seven rejection codes, running three rule sets at once, according to the company's blog post. A regulated sales call, in other words, is checked against more than one kind of rule at the same time:

  • Regulatory obligations the brokerage had to meet
  • The energy suppliers' own contractual terms
  • Industry codes and regional law

The company says the rules are version-controlled and traceable back to the source document that produced them, so a change to a supplier's terms or a regulator's code can be tracked to the clause that changed it.

A hazy daytime view of Dubai's high-rise skyscrapers along Sheikh Zayed Road, seen across an empty sandy lot with scattered desert scrub.
Imre Solt / Wikimedia Commons (CC BY-SA 3.0)CC BY-SA

The company's own performance figures

The figures xAIa publishes for ComplAI are the company's own claims about its own product, not numbers audited or confirmed by a regulator or an independent testing body.

According to the product page, ComplAI reviews 100% of recorded conversations, compared with the roughly 2% a manual team can typically sample, a jump the company itself describes as a 50 times increase in coverage. The company says a cross-check that takes a human reviewer about 45 minutes takes ComplAI about 90 seconds, which it describes as roughly 30 times faster, and that manual compliance review effort fell by more than 80% after the product went live. Beyond the single brokerage example, xAIa's product page states that ComplAI now processes more than 600,000 calls a month across all of its deployments combined. The two figures describe different things and are not directly comparable: the 700,000 above is one deployment's volume in a single month, while this is an open-ended floor the company gives for its current run rate.

xAIa says ComplAI runs on GCC (UAE) data residency with SOC 2-aligned controls and alignment with the GDPR and the UAE's PDPL. The company also states that client data is never used to train shared models. GDPR is described by GDPR.eu as the toughest privacy and security law in the world, applying to any organisation that targets or collects data on people in the EU, regardless of where that organisation is based.

Specifications, percentages and the other figures cited here are as xAIa published them on its blog and product page at the time of writing, and may change as the company updates ComplAI or takes on new deployments and clients.

Both the account of why ComplAI was built and the performance figures attached to it come from xAIa's own blog post and product page. EgyptPulse has not seen an independent audit of those figures, and no third-party or regulatory assessment of ComplAI's accuracy or coverage claims was available at the time of writing. Readers should treat the coverage, speed and cost figures above as the vendor's own account of its product, not as verified outcomes.

Common questions

What is ComplAI?

ComplAI is a product built by xAIa that reads recorded calls, chats and documents against a client's compliance rules and returns a verdict on each one, with the evidence behind that verdict attached, according to the company's product page.

Who owns xAIa?

On its ComplAI product page, xAIa describes itself as "xAIa — a PRO TECHnology Group company" and as "The AI consultancy and implementation arm of PRO TECHnology Group — taking MENA enterprises from AI ambition to AI in live operation."

Are ComplAI's performance figures independently verified?

No. The coverage, speed and effort-reduction figures cited above are xAIa's own published claims about its own product. EgyptPulse has not seen an independent audit or regulator assessment confirming them, and none was available at the time of writing.

What data protection standards does xAIa claim for ComplAI?

xAIa says the product runs under GCC (UAE) data residency, with SOC 2-aligned controls and alignment with the EU's GDPR and the UAE's PDPL, and that client data is not used to train shared models, according to the company's product page.

How did this story make you feel?

Share this story

Follow Us

Written by

Dina Mansour

Writes about health and society — hospitals and public health, the cost of care, education, and the everyday questions readers put to a newsroom about services they depend on. The beat is anchored in what a policy means for a family: waiting times, fees, what is covered and what is not. Health claims are sourced to a medical authority or a published study, never to a press release, and the story says plainly when the evidence is thin.