Press release

Moveo’s LLM vs GPT-4 for Customer Experience

0
Sponsored by Businesswire

Moveo.AI announced that after rigorous comparison, its custom LLM tuned for CX outperforms GPT-4-0613 in all grading dimensions, except Markdown, where GPT-4 performs better. The evaluation was based on a random sample of hundreds of entries from Moveo’s production data, which neither our LLM nor GPT-4 had encountered before. Each entry was converted into a prompt consisting of the user question, conversation history, grounding knowledge from the collection documents, live instructions, and custom instructions.

Advertising
Advertising

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20240723855013/en/

As can be clearly seen in this table, Moveo’s custom LLM outperformed GPT-4 in four critical dimensions that are the cornerstone of a great Customer Experience: Hallucination, Repetitions, Disambiguation, and Readability. The two models are equal in Language while GPT-4 performs better only in Markdown use. (Graphic: Business Wire)

As can be clearly seen in this table, Moveo’s custom LLM outperformed GPT-4 in four critical dimensions that are the cornerstone of a great Customer Experience: Hallucination, Repetitions, Disambiguation, and Readability. The two models are equal in Language while GPT-4 performs better only in Markdown use. (Graphic: Business Wire)

Methodology

The grading process assessed Moveo’s LLM and GPT-4 responses across 8 dimensions that capture critical traits within the CX setting:

  • Hallucination

  • Repetition

  • Disambiguation

  • Live agent handover

  • Readability

  • Language

  • Markdown, and

  • Latency

Each dimension received a score, determining which LLM provided a better response. To evaluate the performance of the different models, Moveo used a separate GPT-4 instance as a “grader,” performing a single API call for each of the samples.

Results

Moveo’s custom LLM outperforms GPT-4-0613 in all grading dimensions, except in Markdown, where GPT-4 performs better in stylistic formatting. Most importantly, it is worth mentioning that in terms of hallucination, GPT-4 performs worse, which could hurt Customer Experience. For example, if GPT-4 provides incorrect information about a product, it could lead to potential liabilities, customer dissatisfaction, and increased support requests.

Moveo’s LLM responds in only 5 seconds, while GPT-4 takes at least 18 seconds. In that time, Moveo.AI could have handled more than 4 inquiries, significantly enhancing support efficiency and customer satisfaction.

According to Panos Karagiannnis, CEO of Moveo.AI, “Enterprises need vertical-specific LLMs as every customer interaction is an opportunity to build trust and loyalty. By minimizing hallucinations and connecting to real-time information systems, our LLM significantly beats GPT-4, reduces the risk of customer dissatisfaction and potential liabilities, and sets a new standard in CX”.

To learn more about Moveo’s proprietary LLMs, please visit: https://moveo.ai/

About Moveo.AI

Moveo.AI is a Conversational AI platform transforming how enterprises interact with customers. Moveo’s LLM, trained on historical and real-time CX data, powers GenAI agents to seamlessly connect to real-time data and unstructured knowledge bases to provide accurate and contextually relevant answers to inquiries.