- The MIT State of AI in Business 2025 report shows that companies have spent $30-40 billion on generative AI, yet 95% have not seen a return on investment; only 5% have crossed the “GenAI Divide.”
- Qlik & ESG: 94% of companies are increasing AI investment, but only 21% are operating effectively. Informatica: The main causes are poor data quality, lack of readiness, and immature infrastructure.
- The problem lies not in the AI models themselves, but in the lack of integration, the inability to measure data fitness, employee expectations, and dynamic governance.
- Traditional frameworks (RICE, ICE, MoSCoW) fail with AI because:
- Reach is based on absolute user numbers, which is easily exaggerated.
- Confidence is subjective, ignoring data risks and model capabilities.
- Effort only accounts for coding, not the cost of data cleaning.
- Impact is hard to measure because AI can either assist or replace humans, and behavior is inconsistent.
- Apple, with its “The Illusion of Thinking” research, highlights the limitations of Large Reasoning Models (LRMs): difficulty generalizing beyond training data and unstable behavior.
- A 2025 Stanford study of 800 tasks/100 professions found that employees want AI to assist with nearly 50% of their work, but many AI projects focus on the wrong tasks. The proposed “Human Agency Scale” shows the highest value lies in AI assistance, not replacement.
- ARISE (AI Readiness and Impact-Scoring Evaluation) was created to replace RICE:
- It keeps Reach, Impact, Confidence, and Effort but standardizes their scales.
- It adds 3 new factors: AI Desire (real user need), AI Capability (data, model maturity), and Intent (AI assistance vs. automation).
- Formula: ARISE Score = (Reach * Impact * Confidence / Effort) × AI Desire × AI Capability × Intent Multiplier.
- Example: An AI coding assistant (ARISE score = 20) is prioritized higher than an automated bug-fixing AI (score = 1) because data capability is still weak.
- ARISE helps businesses avoid chasing shiny demos and focus on projects with real value, balancing humans and AI.
📌
Despite $30-40 billion being poured into generative AI, 95% of organizations have yet to reap the benefits, primarily due to outdated project management frameworks and a lack of data readiness. ARISE emerges as an AI-native tool to replace RICE, adding metrics for AI Desire, Capability, and Intent to ensure feasibility and real-world value. This helps businesses avoid common pitfalls, prioritize the right AI projects, and balance human augmentation with automation.
