Baidu, Inc. Q2 FY2025 earnings call

• AI cloud: Revenue reached RMB 6.5 billion in Q2, up 27% y/y. Non-GAAP operating profit grew y/y. Subscription-based revenue in enterprise cloud grew solidly. Achieved large-scale stable deployment of prefiled detailed separation architecture, improving inference and reducing costs. • Autonomous driving: Apollo Go provided over 2.2 million fully driverless rides in Q2, up 148% y/y. Global partnerships with Uber and Lyft, expanding into international markets. Made progress in Hong Kong and Middle East. RT6, a purpose-built Level 4 autonomous vehicle, provides cost efficiency. • Mobile ecosystem: Accelerated AI transformation of search. AI-generated content reached over 50% of mobile search result pages by end of June, up from 35% in April. Baidu App MAU reached 735 million in June, up 5% y/y. Daily average time spent per user in Q2 increased 4% y/y. Began prudent small-scale monetization testing in Q2.

Baidu Core's total revenue was RMB 26.3 billion in Q2. AI cloud revenue was RMB 6.5 billion, up 27% year-over-year, contributing a significant portion to Baidu Core's non-online marketing revenue which exceeded RMB 10 billion for the first time, up 34% year-over-year. Revenue from iQIYI was RMB 6.6 billion, decreasing 11% year-over-year. AI cloud revenue of RMB 6.5 billion represents a notable portion of Baidu Core's non-online marketing revenue growth.

• Continue accelerating foundation model integration, focusing on application value areas. • Working on ERNIE 5.0 with significant improvements across key capabilities, expect to launch when ready. • Continue rolling out iterations and updates on existing models. • Monitor industry developments to align technology roadmap with market opportunities.

• Market competition: Intensifying competition in AI cloud and autonomous driving sectors. • Chip supply: Easing of H20 chip restrictions but need to monitor supply environment and ensure flexible AI architecture to support various chips. • Monetization challenges: AI search monetization still in early stages, near-term pressure on revenue and margins as transformation continues.

Q: How to assess sustainability of AI-driven cloud demand and chip constraints?

A: Strong and growing demand for AI-driven cloud services across sectors. Differentiated by cost-effective end-to-end AI cloud products due to full stack AI capabilities. Focus on building flexible AI architecture to support various chips, including domestic ones, to adapt to evolving supply environment.

Q: How does Apollo Go assess long-term differentiation and Uber partnership?

A: Apollo Go is a global leader in autonomous driving. Has cutting-edge technology, massive investment, and disciplined execution. Partnerships with Uber and Lyft help expand into global markets, leveraging partners' local presence. Focus on running real-world operations at scale and achieving sustainable profitability through cost efficiency and global expansion.