However, not all industries have found value in AI due to concerns about cost and risk. To avoid disillusionment with AI investments, the author suggests developing an AI policy, using specialized AI tools to reduce risk, and using specialized models for bigger benefits. For example, intelligent document processing (IDP) can autonomously process certain document types, reducing ethical risks and furthering business objectives. The author concludes that by narrowing the scope of AI initiatives with specialized models, organizations can set achievable and measurable goals.
Key takeaways:
- Despite high expectations, the actual ROI from AI investments has fallen slightly short, with 47% of respondents indicating that their actual ROI reached their expectation of at least twice the cost of their investment.
- More than 50% of businesses are expected to abandon their proprietary large language model (LLM) projects by 2025 due to factors like cost, complexity or technical debt.
- AI has been widely adopted by developers and the financial services industry, with use cases including predictive analytics, code generation, data extraction and analysis, and performance analysis.
- To avoid disillusionment with AI investments, businesses should develop an AI policy, use specialized AI tools to reduce risk, and opt for specialized models that offer robust utility in more specific areas.