- Operator inventory systems are less than 50-60% accurate, leaving most carriers without the data foundation AI-driven operations require
- Most operators sit at Level 2-3 network autonomy, several levels short of what the AI opportunity demands
- Legacy OSS/BSS systems and the organizational inertia that cost telcos the cloud era haven't gone away, according to the latest Fierce Network Research report
Telcos own an essential asset for the AI economy. Whether they can exploit that asset is the question. A new Fierce Network Research report identifies four specific barriers standing in the way.
The asset is proximity. AI training runs in centralized hyperscale data centers, but inference — AI doing real work in the world — must run close to users and data. Telcos are already where inference needs to be, with towers, fiber and edge facilities distributed across every market.
"Our infrastructure being closer to the end user, much closer than the hyperscalers, gives us the license to play and win in this economy," said Salim Kouidri, SVP of field engineering at T-Mobile, in an interview for our report, "AI and the automated network: Designing telco infrastructure for the age of inference," makes that case in detail. It also catalogs what's in the way.
Barrier one: Dirty data
Most operators' inventory systems are less than 50-60% accurate, said Gabriele Di Piazza, VP of product management at Blue Planet. Operators cannot trust data they cannot verify — which means the foundational layer for AI-driven operations is simply missing at most carriers. Automation built on bad data produces wrong answers at machine speed. Gartner research reinforces the point: 38% of organizations that experienced AI setbacks cited poor data quality as a direct cause.
Barrier two: The autonomy gap
Most operators classify themselves between Level 2 and Level 3 on the network autonomy scale, Di Piazza said — an assessment TM Forum's March report confirms. The AI opportunity requires networks that interpret intent and act autonomously, with humans supervising rather than executing. That's Level 4 and above. Most of the industry is several levels short.
Barrier three: Organizational inertia
Telcos are traditionally slow-moving environments, and for defensible reasons — five nines of reliability demands caution. But legacy OSS and BSS systems written in archaic languages constrain agility, and the organizational inertia that caused telcos to miss the cloud transition has not disappeared.
"They've messed up the whole cloud opportunity," said Sid Nag, president and chief research officer at Tekonyx. "And if telcos want, this is their second chance."
Barrier four: The consumption gap
Hyperscalers won the cloud era partly by making services effortless to buy and provision. Carriers still can't match that. "Carriers have been aspiring to that capability for quite some time, but they're not to the point where they can deliver as hyperscalers do," Di Piazza said.
None of these barriers is permanent — operators like T-Mobile and MetTel are already demonstrating what clearing them looks like, with MetTel's AI engine making analysts 83% more efficient in some years. But the report's implicit warning is clear: the asset alone isn't enough. The operators that fix the data, climb the autonomy ladder and move faster than their culture prefers will capture the value. The rest will watch it flow past them — over their own networks.
Download the full report, "AI and the automated network: Designing telco infrastructure for the age of inference,", and watch the companion virtual summit on demand: AI and the Automated Network.