Insights from AI Surveys

For the last 9 months, I am leading an AI incubation project that leverages GenAI capabilities to enhance customer response quality and resolution efficiency. We faced things we had never dealt with before. It was all new! The most prominent challenges were — Defining a clear use case, selecting the right LLM & the relevant AI partner, data privacy clarity on the training dataset, and the most significant one among all is the “hallucinations (response quality)”. We have come a long way by navigating these challenges.

I am keen to compare my own project’s challenges & opportunities with broader trends in the AI field. There is a good number of surveys conducted to provide insights on diverse perspectives on the impact of AI in Enterprises, Open source, and Leadership levels.

Looked at a bunch of surveys — OReilly, Linux Foundation, KPMG & Deloitte. Following are the voices from the survey that resonate with me. Figure out how many of these are relevant for you.

• Most of the organizations surveyed believe Generative AI will have the largest impact on their businesses out of all emerging technologies. majority of companies surveyed plan to significantly invest in GenAI, allocating a large percentage of their IT budgets to the technology.
• Survey results paint an impression that improvement in workforce productivity is the biggest factor for investing in AI. Across organizations, their current generative AI efforts remain more focused on efficiency, productivity and cost reduction than on innovation and growth.
• Many AI adopters are still in the early stages and most of them are in the POC phase for less than a year. While, some of them already have applications in production. However, most of them plan to rollout their first generative AI solution within the next two years.
• Most organizations are primarily relying on off-the-shelf generative AI solutions & roughly 20% of the organizations are working with AI are using open source models.
• The prominent adoption challenges highlighted across the survey results were, Lack of clear business use cases, Lack of talent, Governance, Cyber Security & Data Privacy. Organizations and Leaders are looking for more regulation and collaboration globally.
• Most prominent implementation challenges highlighted are, i) Unexpected outcomes (Response Quality), ii) Security vulnerabilities, iii) Safety & Reliability, iv) Fairness, Bias & Ethics, and Privacy are the significant areas for which adopters are testing.
• Prompt injection & Model Leeching are the emerging vulnerabilities in generative AI with growing concerns that require further research and development of robust security solutions
• Open-source advocates believe that open-source GenAI tools are ultimately more sustainable in the long run compared to closed-source, proprietary options.
• The legal consequences of using generative AI are still unknown. This also poses a challenge in terms of who can own the copyright.

Originally published at https://www.linkedin.com.