5G Factor: AI RAN and Telco AI Rising

In this episode of The 5G Factor, our series that focuses on all things 5G, the IoT, and the 5G ecosystem as a whole, we look at the recent key AI RAN and telco AI moves, especially alliance formations unveiled before and during Mobile World Congress 2024. The major AI RAN and Telco AI developments include the near-term and long-term ecosystem impact of the AI-RAN Alliance launch, HEAVY.AI's progress in the key GenAI areas of accuracy and speed, the official debut of the Global Telco AI Alliance (GTAA), and the AI Alliance's focus on responsible AI.

Our analytical review focused on:

AI-RAN Alliance Debuts: How Substantive? The AI-RAN Alliance debuted at MWC 2024, a new collaborative initiative aimed at integrating AI into cellular technology to further advance RAN technology and mobile networks. The alliance's founding members include AWS, Arm, DeepSig, Ericsson, Microsoft, Nokia, Northeastern University, NVIDIA, Samsung Electronics, SoftBank and T-Mobile. The group's mission is to enhance mobile network efficiency, reduce power consumption, and retrofit existing infrastructure, setting the stage for potentially unlocking new economic opportunities for telecommunications companies with AI, facilitated by 5G and further out 6G. We examine how network operators in the alliance are set to spearhead the testing and implementation of these evolving technologies developed through the collective research efforts of the member companies and universities and ecosystem prospects for AI RAN for both near-term and long-term.

HEAVY.AI Showing Accuracy and Speed Breakthroughs. HEAVY.AI announced HeavyIQ designed to bring LLM capabilities to the GPU-accelerated HEAVY.AI analytics platform, with the goal of enabling organizations to interact with their data through conversational analytics. Users can explore their data with natural language questions and generate advanced visualizations of that data. This streamlined process could reduce the friction of traditional business analytics, allowing more users to swiftly uncover insights. With HeavyIQ, HEAVY.AI has taken an open-source LLM foundation model and extensively trained it to perform core analytics tasks, including analyzing massive geospatial and temporal data sets. The technology employs LLM in conjunction with retrieval augmented generation (RAG) capabilities to take a user's text input, automatically convert it into a SQL query, and can both visualize and return natural language summaries of results. We review why HEAVY.AI is providing benefits in areas such as accuracy. Trained with over 60,000 custom training pairs, benchmarks show HeavyIQ to be more accurate than GPT-4, with 90%+ accuracy on common text-to-SQL benchmarks, compared to 85% with GPT-4. For speed, HeavyIQ leverages optimized and fine-tuned smaller models that take advantage of the latest NVIDIA GPU hardware innovations to deliver responses up to 10x faster than GPT-4.

Global Telco AI Alliance Officially Takes Off. Deutsche Telekom, e& Group, Singtel, SoftBank and SK Telecom officially launched the Global Telco AI Alliance (GTAA), following its pre-announcement in July 2023. The telcos also announced plans to establish a joint venture, through which the companies plan to develop LLMs specifically tailored for the requirements of telecommunications companies. The LLMs will be designed to help telcos improve their customer interactions through digital assistants and chatbots. We explore why it is important to the mobile ecosystem for the JV to develop multilingual LLMs optimized for languages including Korean, English, German, Arabic and Japanese, with plans for additional languages to be agreed among the founding members and advancing telco specific LLMs that align to the telecommunications domain and can prove better at understanding user intent.

AI Alliance Prioritizes Responsible AI. In December 2023, IBM and Meta launched the AI Alliance in collaboration with over 50 founding members and collaborators. The AI Alliance is focused on cultivating an open community and enabling developers and researchers to accelerate responsible innovation in AI while ensuring scientific rigor, trust, safety, and economic competitiveness. By bringing together developers, academics, scientists, and other innovators, the alliance looks to pool resources and knowledge to address safety issues. We assess why the AI Alliance is integral to fulfilling ecosystem-wide objectives including deploying benchmarks and evaluation tools and standards to enable the responsible development and use of AI systems (i.e., Responsible AI). This includes supporting AI skills building and exploratory research.