Essential Lessons for Business Leaders in the Era of AI Large Models
Alibaba Cloud Developers — 2025-10-23 (Zhejiang)
Speakers:
- Jiang Linquan — Vice President & CIO, Alibaba Cloud Intelligence Group
- Liu Xiangming — Co-founder & Co-CEO, TMT Media
Both are seasoned practitioners with extensive engagement in enterprise AI. They shared candid insights on real-world challenges in AI implementation, offering a practical guide for decision-makers.
---
Conference Session Overview
Topic: AI Large Model Era: Essential Lessons for Executives
Focus Areas:
- How the CIO role is evolving in the AI era
- Defining AI-driven business value
- Identifying high-potential application scenarios
- Crafting & benchmarking enterprise AI strategies
- Overcoming adoption barriers
- Achieving organizational cognitive alignment
---
Jiang Linquan’s “Five Phases” of Enterprise AI Adoption
- Collective Calm
- Localized Excitement
- Systemic Pressure
- Landing Obstacles
- Cognitive Alignment
Key takeaway: The pivotal question is—What can AI actually do?
---
Liu Xiangming’s Silicon Valley Insight
> "If software already does it well, AI shouldn't do it."
Jiang Linquan’s perspective: AI should enhance mature systems to lower barriers for non-technical staff—like adding cherries to a well-baked cake.
---
Q1: CIO Role Evolution in the AI Era
Jiang Linquan:
- Information ➡ Data ➡ Intelligence progression.
- AI can now fully utilize unstructured “language” data in contracts, SOPs, customer service scripts.
- Regardless of focus area, CIO’s core mission: use digital tech to transform workflows into valuable insights.
Liu Xiangming:
- Shift from convincing leaders to use computers ➡ managing demands to implement AI.
- Requires new skills, broader collaboration, and expectation alignment.
- Key advice: Be the CIO who uses AI best—equip yourself with knowledge, skills, and resources.
---
Q2: Defining Business Value & Choosing AI Scenarios
Jiang Linquan — Three Criteria for AI Entry Points:
- Language-centric tasks
- Customer service, telesales, contract review, internal knowledge bases.
- Repetitive processes suitable for batch execution.
- Workload pressure and efficiency needs.
Liu Xiangming:
- Get leadership personally using AI to form genuine understanding.
- Direct interaction builds realistic perception.
---
Q3: Setting AI Strategy & Measurement Benchmarks
Jiang Linquan:
- AI fits best into tasks with clear SOPs and measurable metrics.
- Define problems from actual demand—measurement becomes straightforward.
Liu Xiangming:
- AI can reduce interpersonal friction by unlimited iteration without emotion.
---
Q4: Real-world AI Practices & Breakthrough Scenarios
Jiang Linquan:
- Added AI Chatbot to Alibaba Cloud site—10× efficiency over search, but quality issues emerged.
- AI is more demanding than search—requires strong supply-side capabilities and user skills.
Liu Xiangming:
- Effective prompts and understanding of keywords are crucial.
---
Q5: Biggest Resistance to AI Landing in Enterprises
Liu Xiangming:
- Primary barrier: cognitive gap and uneven information diffusion.
- Anxiety-driven decision making in executives.
Jiang Linquan:
- Those with deeper AI experience resonate more with implementation challenges.
Liu Xiangming:
- 80% efficiency gains so far from organizational/process optimization, not AI itself.
- AI acts as a mirror revealing systemic issues.
---
Q6: Private vs. Open-source Models
Jiang Linquan:
- No unified view; selection requires measurement within business scenarios.
- Use SOTA models only if measurement capability exists.
---
Q7: Aligning Business & Technical AI Perceptions
Mechanisms:
- Cross-functional workshops
- Joint pilot projects
- Shared performance metrics
Jiang Linquan:
- Internal AI certification programs align understanding across departments.
- Examples: ACA & ACP certifications for large models.
Liu Xiangming:
- CIOs must act as chief evangelists—spreading vision and driving adoption.
---
Q8: Advice for CIOs Starting AI Transformation
Jiang Linquan — Five Stages of Collaboration:
- CEO & CIO remain calm
- CEO excitement mobilizes team
- Systemic execution pressure
- Practical challenges force reality check
- Balance excitement & calm—AI fully lands
Success Metrics:
- AI upgrades boosting accuracy by 3%
- Cost reductions of 20–30×
- Continuous evolution once embedded into workflows
Recommendations:
- Initiate full AI strategy immediately
- Begin hands-on trials and closed-loop iterations today
Liu Xiangming:
- Focus on people alignment and core scenario selection
- Learn through trial and error—action overcomes fear
---
Integration Tip: Use Platforms for Practical AI Deployment
Example: AiToEarn — an open-source global AI content monetization platform:
- AI content generation ➡ cross-platform publishing ➡ analytics ➡ model ranking (AI模型排名)
- Supports publishing to Douyin, Kwai, WeChat, Bilibili, Xiaohongshu, Facebook, Instagram, LinkedIn, Threads, YouTube, Pinterest, X (Twitter).
- Ideal for rapid AI iteration and measurable impact.
Resources:
---
Final Takeaway:
AI transformation demands aligned cognition, targeted scenario selection, and iterative practice. CIOs must lead from the front—educating, experimenting, and embedding AI where it delivers real, measurable value.
---
Do you want me to also create infographic-style summary tables for each Q&A section so this Markdown becomes visually skimmable? That would make it even easier to present.
