Sea Limited is deepening its relationship with Google, signing a memorandum of understanding to advance AI across Shopee (commerce), Garena (gaming), and Monee (digital financial services).
While the announcement reads like a standard partnership update, the specific areas named (agentic commerce, agentic payments, and AI-assisted game operations) signal something more pointed: both companies want AI systems that don’t just recommend, but act.
That distinction matters. Most consumer AI in 2024-2025 was about chat and content. “Agentic” systems are about execution: software that can navigate interfaces, compare options, apply rules, and complete transactions with minimal human input. Done well, it removes friction.
Poorly done, it becomes an expensive layer of confusion, or worse, a security liability.
Also Read: Agentic AI is powerful – but power isn’t product-market fit
This expanded partnership builds on existing Sea-Google collaborations, such as the YouTube Shopping Affiliate Program with Shopee and Free Fire League on Google Play with Garena. The new element is explicit: Sea wants to operationalise AI at scale across its ecosystem, and Google wants distribution in some of the world’s most mobile-first markets.
Forrest Li, Sea’s Chairman and CEO, framed it as the next platform shift:
“At Sea, we have always believed in the fundamental power of technology to improve lives and create long-lasting value for the communities we serve.
AI is the next big technology revolution, and we believe that it has huge potential to positively transform our business and create value in our markets. This partnership with Google on AI will drive innovation in the business application of the technology at scale, and allow us to make AI more accessible to the digitally underserved in our markets,” Li added.
Google APAC President Sanjay Gupta struck a similar tone. “By combining Google’s AI leadership with Sea’s innovative ecosystem, we’re building products that don’t just solve today’s challenges but define the future of gaming, commerce, and financial services. And we’re developing these solutions responsibly, with user privacy and safety at the core. Together, we are accelerating the adoption of this transformative technology and unlocking the immense economic potential of Southeast Asia’s digital landscape.”
The real story is what “agentic” implies for shoppers, gamers, and people who still sit on the edge of formal finance.
1) AI-powered agents and the future of online shopping: from browsing to delegation
Shopee and Google say they will “jointly explore the building of an AI agentic shopping prototype” that “can seamlessly integrate across Shopee and Google platforms”.
If that prototype becomes a product, it could shift online shopping from search-and-scroll to goal-driven delegation. Instead of a user typing keywords, filtering, opening ten tabs, checking delivery dates, and messaging sellers, an agent could:
- Translate a vague intent (“cheap, reliable phone for my mum”) into specific constraints
- Compare sellers, shipping times, warranty terms, and return policies
- Watch prices over time, alert on drops, and execute purchases within a budget
- Bundle items to optimise shipping or apply the proper vouchers automatically
- Handle post-purchase steps: tracking, rescheduling delivery, filing returns
Southeast Asia is an unusually fertile market for this because commerce is already messy in the real world: multiple languages, informal sellers, heavy promotion mechanics, and a wide range of logistics reliability. An agent that can actually navigate those trade-offs could become the new front door to shopping.
Also Read: Why agentic AI isn’t what the hype suggests
But it also raises uncomfortable questions. Who does the agent really serve — buyer, seller, or platform margin? If an AI agent becomes the shopping interface, then ranking, sponsored placements, and “recommended” choices become even more consequential. Platforms will need to show that agents are not simply optimised to maximise take rate while wearing a friendly chatbot mask.
2) AI innovation in gaming: not just smarter NPCs, but faster live ops and globalisation
On the gaming side, Garena and Google are looking to use Google’s AI solutions to “enhance gamer experiences” and “transform the productivity of game development and operations”, with a line about early access pilots for Google’s latest AI research.
The obvious consumer-facing play is richer worlds: better non-player characters, more adaptive matchmaking, personalised onboarding, and dynamic content. The less glamorous—but likely more valuable—angle is operations:
- Faster content production (assets, localisation, event scripting) to keep live-service titles fresh
- Better moderation and trust-and-safety tooling for voice and chat, where toxicity kills communities
- Anti-cheat systems that can detect novel patterns rather than just known signatures
- Smarter A/B testing loops that tune difficulty and retention without breaking fairness
If Garena can shorten content cycles and improve trust and safety, it can scale globally with less operational drag. That matters because “global gaming experience” is often less about graphics and more about whether a game feels fair, stable, and culturally native in Bangkok, Manila, São Paulo, and Riyadh at the same time.
There’s a catch: generative tooling can also turbocharge bad actors—cheat creation, scams, and automated harassment. Any AI advantage in gaming will be matched by AI-powered abuse, and publishers will have to budget for that arms race.
3) AI and financial inclusion: fewer forms, more approvals, but also new kinds of exclusion
Sea’s financial arm, Monee, will work with Google on an “open, shared Agent Payments Protocol (AP2)”, where Monee will provide feedback to ensure it is “robust, secure, and suitable” for Southeast Asia, with an intention to later explore pilot experiences across platforms.
If AP2 evolves into something widely adopted, it could reduce one of the biggest blockers to financial inclusion: complexity. Many underbanked users don’t struggle with the idea of digital money; they struggle with onboarding steps, confusing user interfaces, and customer support that doesn’t speak their language or understand their context.
AI could help by:
- Turning onboarding into a guided, multilingual flow that adapts to user capability
- Automating dispute handling and customer service at lower cost
- Improving fraud detection to protect first-time users (who are prime scam targets)
- Enabling small merchants to accept digital payments and reconcile accounts without accounting expertise
For SMEs, inclusion is not philosophical; it is operational. If agents can reconcile transactions, chase invoices, or manage cashflow nudges, that’s not “AI magic”; it is time returned to a shop owner.
Still, AI-driven finance comes with a risk the industry often underplays: automated denial. Models can quietly exclude people with thin files, unstable device histories, or non-standard income patterns; the exact users inclusion efforts claim to prioritise. Any “agentic payments” system that touches identity, fraud, or credit will need strong controls, auditability, and clear recourse when automation gets it wrong.
Also Read: Agentic AI: The next frontier in technology
Sea and Google are calling this partnership “strategic”. The test will be whether these projects become everyday tools that work in the region’s real conditions: inconsistent connectivity, diverse languages, scam-heavy environments, and users who will not tolerate extra steps just because the system is “smart”. Agentic AI only wins if it makes life simpler — and doesn’t create a new category of problems that humans then have to clean up.
The post Shopee, Garena, Monee: Sea’s AI ambition gets serious appeared first on e27.