From dial-up to AI: Unlocking Telecom’s next frontier while navigating AI regulatory pitfalls

As mentioned in the introduction, the only constant in the telecom industry is the development of technology and the continuous implementation of the latest technological advancements. This is simply the way telecom companies must operate to remain competitive.

Let’s take a look at some technological eras that have shaped telecoms in recent decades.

Dial-Up Internet (1980s - 1990s) - The First Digital Revolution in Telecom

In the early 1980s, making a voice call was the primary need for end users, while the internet was something that existed but was far from mainstream use. This period was marked by very low data transfer speeds (typically below 50 kbps), and the internet was both expensive and unreliable.

However, the demand for internet-based services started growing, presenting the first major challenge for telecom companies in the digital era, which had to transition quickly to broadband and move away from the slow and unreliable dial-up technology.

Broadband & ADSL (2000s) - Unprecedented Speed

With the arrival of ADSL, everyone was talking about the internet, and by the mid-2000s, the .com bubble had reached its peak. The internet became significantly faster, and users no longer had to choose between making a phone call or using the internet - they could do both simultaneously, which was revolutionary at the time.

Broadband enabled streaming, online gaming, the emergence of the first cloud-based platforms, and the rapid development of internet technologies.

Recognizing the potential of internet technology, telecom companies faced a new challenge. Data traffic became more profitable than voice calls - so how could they shift users toward internet-based communication while moving away from traditional communication methods?

Telecoms increasingly transitioned from metered data plans to flat-rate packages to encourage maximum internet usage, from messaging to voice calls. New investments were made in fiber-optic cables and the expansion of mobile devices as the primary medium for internet access.

The Era of Smartphones and 4G (2010s) - Mobile Internet Became the Standard

With the technical requirements met by integrating 4G technology into mobile devices, phones became smart devices, allowing users direct access to the internet. This marked a turning point in how the internet is used today, as most services shifted to being provided online, directly on users’ mobile devices.

From messaging apps like WhatsApp, Viber, and Telegram to navigation services such as Google or Apple Maps, and entertainment platforms like YouTube, Netflix, and Spotify.

However, technological advancements and the evolving expectations of users demanded even more from telecoms, particularly in terms of service versatility and speed.

5G Technology (2020s) - More Than Just Connectivity

Unlike previous eras, where the focus was on service speed, the development of 5G technology had an unimaginable goal - network slicing, ultra-low latency, and massive IoT connectivity.

5G technology pushed telecom companies beyond the traditional telecommunications industry by offering platform-based services to their customers, such as private 5G networks, smart cities, and industrial automation.

Although it required massive investments in infrastructure, this era laid all the technical and business foundations for the next phase of development - AI.

As we have seen above, each transition and advancement in the telecom industry has not only meant increased speed or service quality but also a shift in how telecom companies operate and deliver the value to their users:

• Dial-up → Broadband - shift from voice-first to data-first
• Broadband → 4G - shift from home internet to mobile internet
• 4G → 5G - shift from mobile internet to universal connectivity

The next shift represents a move from universal connectivity to artificial intelligence. This advancement is not just a regular upgrade. It is a transformation in how networks function, how services are delivered to users, how users interact with telecom companies, and ultimately, how telecoms provide value to their customers.

We have previously discussed how telecoms are leveraging AI and the risks associated with it in our blog post The AI revolution in telecom: Maximizing benefits while mitigating risks.

However, in the following section, we will explore the latest challenge facing telecom companies - one that is not tied to the technology they develop themselves: AI regulations.

As we have already mentioned, telecom companies are leaders in adopting the latest technological advancements. Therefore, AI adoption in telecom is not a future event - it is something we are already witnessing. A quick visit to telecom companies’ websites reveals that almost all of them have already implemented customer-facing AI solutions, such as AI customer support chatbots.

This means telecoms have already widely adopted AI. However, with great opportunity comes great responsibility. Governments and regulators worldwide are aware of this and are actively working on legal frameworks to regulate the development, deployment, and use of AI. The EU AI Act is the first law of this kind, and you can learn more about it in our guide.

For telecom companies, this introduces a new layer of responsibility. Unlike previous technologies they have adopted, AI is far more advanced, processes vast amounts of data, and carries significantly higher risks. As a result, regulatory focus is increasing, and the EU AI Act serves as the best example of legally mandated responsibility that comes with AI adoption.

Before we move forward, let’s explore how the EU AI Act affects telecom companies.

How Does the EU AI Act Regulate AI Use in Telecom Companies?

As detailed in our guide, the EU AI Act is currently the only comprehenisve regulation in the world that regulates the development, deployment, and use of AI. It categorizes AI systems based on their level of risk, classifying them into:

• Unacceptable risk - AI systems that manipulate behavior and cause clear harm to society, which are outright banned.
• High risk - AI systems that can make or impact significant decisions about people and users and/or are implemented in critical infrastructure, healthcare, education, law enforcement, transportation, or in employment context.
• Limited risk - AI systems that do not pose significant risks but still interact with users.
• Minimal risk - AI systems that do not present any serious risks.

Since telecom networks are usually considered critical infrastructure, a signigicant part of AI used for network management, optimization, fraud detection, and other telecom applications likely falls under the high-risk AI category. 

This means that such AI systems require detailed documentation, risk assessments, and continuous human oversight to ensure compliance and minimize potential risks.

Telecom Companies Are Responsible, Even If Their AI Vendors Are Not Compliant

Now that we understand how the EU AI Act classifies AI used in critical infrastructure such as telecom networks, let’s discuss what this means for telecom companies.

In short, one of the biggest pitfalls for telecom companies could be an assumption that compliance is solely the responsibility of their AI vendors (such as cloud providers or AI software suppliers). The reality is different.

According to the EU AI Act and Product Liability Directive, telecom companies need to contractually make sure that providers of their AI are meeting  regulatory compliance demands, and make sure that they are compliant as well, as AI deployers.,

For telecom companies, this specifically means:

• If AI systems used for network optimization or fraud detection are not compliant with the EU AI Act, the telecom company is held responsible.
• If a customer-facing AI chatbot does not meet EU AI Act requirements, the telecom company is liable.
• If an AI model managing network forecasting fails and causes service outages, the telecom provider could be held accountable for damages.

In short, simply implementing AI to increase efficiency and achieve business goals is not enough. Telecom companies must ensure that their AI vendors’ solutions are not just the best or most cost-effective but also fully compliant with legal regulations.

Now, let’s explore how telecom companies can select a compliant AI vendor.

We understand that telecom companies are massive organizations with a great responsibility toward society. This leaves very little room for risk, and as we’ve seen in this article, regulatory frameworks place a significant burden on telecoms by shifting responsibility from AI vendors to telecom companies.

That’s why it’s crucial to choose an AI vendor wisely - not only ensuring that they offer high-quality AI solutions but also that they prioritize compliance with legal regulations.

The Challenge of AI Vendor Evaluation

The traditional vendor evaluation process is well known - asking a long list of specific questions, waiting for responses, and analyzing the answers. While this method is time-consuming, it does help in selecting a compliant vendor (at least at the time of signing the contract).

However, legal regulations evolve rapidly, and AI technology advances even faster. Regulators expect AI solutions to remain compliant at all times. This means vendor evaluations must be conducted continuously (in real time), making manual assessment a slow and ineffective approach.

A Smarter Way to Evaluate AI Vendors

At TrustPath, we understand the complexity of this process. That’s why we offer a fully automated solution for evaluating and selecting the best AI vendors. With our platform, you can find AI vendors that meet the highest standards in:

• Compliance with legal frameworks (EU AI Act, GDPR, and future regulations)
• AI Model Transparency for responsible AI adoption
• Security & Data Protection to prevent misuse
• AI Model Auditing to ensure accountability

Our automated solution helps you choose the right AI vendor, save procurement time, and adopt AI with confidence.

Contact us to evaluate your AI vendors and protect your business from potential risks!