AI产业跨越式发展的策略思考

Core Viewpoint - The report emphasizes the need for Guangzhou to enhance its AI industry development strategies to improve competitiveness and address existing shortcomings in the sector [1] Group 1: Current State of AI Industry in Guangzhou - Guangzhou has made initial achievements in AI industry development, with numerous companies established in areas such as computer vision, natural language processing, and intelligent robotics [1] - However, the city faces challenges including insufficient industry competitiveness, a lack of leading AI enterprises, high computing costs, and inadequate policy support [1] Group 2: Strategies for Reducing Computing Costs - The introduction of a "computing power voucher" policy is suggested, providing subsidies of 30%-50% based on actual computing power consumption for companies using public computing platforms [2] - Focused support for key areas such as AI chip development and large model training is recommended to help reduce training costs and accelerate model optimization [2] Group 3: Enhancing Technological Autonomy - Increased support for companies using domestic AI chips and algorithm frameworks is proposed, with R&D subsidies ranging from 15%-25% based on usage [3] - A procurement policy prioritizing domestic AI products for government projects is suggested, aiming for at least 80% procurement of domestic products in major projects [3] Group 4: Promoting Application Scenarios - The establishment of an annual AI application scenario list is recommended, encouraging companies to propose solutions and compete for funding and support [4] - Successful solutions should be promoted citywide, with incentives for companies adopting these solutions [4] Group 5: Tax Incentives for Innovation - Recommendations include increasing the R&D expense deduction ratio for key AI sectors and providing tax exemptions for fixed asset investments in AI [5] - Personal income tax reductions for high-end AI talent and support for housing and education are also suggested to strengthen talent retention [5]