Apple Unveils Revamped Siri AI Powered by Google's Gemini at WWDC 2026

By Kim Seong Hyeon Posted : June 9, 2026, 15:09 Updated : June 9, 2026, 15:09
Attendees watch presentations at Apple's annual Worldwide Developers Conference (WWDC) held in Cupertino, California, on June 8. [Photo=Yonhap News]


Apple's WWDC 2026 Features Major Siri AI Overhaul with Google Gemini
 
Apple kicked off its annual Worldwide Developers Conference (WWDC) 2026 on June 8 in Cupertino, California, marking CEO Tim Cook's final keynote address. The company unveiled a significantly revamped Siri AI, which operates on Google's Gemini model.
 
This year's WWDC served as a platform for Apple to overhaul its AI strategy, which has lagged behind competitors. The updated Siri AI features a 3D visualization interface that activates with user gaze and automates tasks across multiple apps, representing the largest update since its launch. However, regulatory risks remain apparent. Due to the EU's Digital Markets Act (DMA), Siri AI will not be available in the European Union and China at the time of the iOS 27 release. Additionally, full functionality will only be supported on iPhone 17 Pro, Pro Max, and Air models, which could impact its broader adoption.
 
Google Signs $920 Million Monthly AI Compute Deal with SpaceX
 
Google has entered into a $920 million monthly AI compute leasing agreement with SpaceX. Under the terms, Google will have access to approximately 110,000 NVIDIA GPUs and CPUs, along with memory resources, from October 2026 through June 2029, totaling around $30 billion.
 
Google described the deal as a short-term contract to secure bridge capacity due to unexpectedly high customer demand for its GemAI Enterprise Agent platform. This agreement follows SpaceX's previous $1.25 billion contract with Anthropic for the Colossus1 data center, marking the second major AI compute deal for SpaceX. Combined, these contracts position SpaceX to secure approximately $75 billion in future contract revenue as it prepares for a Nasdaq IPO roadshow on June 12.
 
The irony of Google, one of the largest AI infrastructure providers, leasing GPUs from a rocket company underscores the current surge in global AI compute demand.
 
SpaceX Aims for Record IPO, Reassessing Value as an AI Cloud Company
 
As SpaceX prepares for its Nasdaq listing on June 12, its value as an AI infrastructure company is being reevaluated. According to IPO documents, xAI spent $12.7 billion on AI infrastructure in 2025 and invested $7.7 billion in the first quarter of 2026 alone. SpaceX plans to raise approximately $75 billion by offering shares at $135 each, which, if successful, would surpass the IPO record set by Saudi Aramco.
 
On June 8, SpaceX also unveiled its first-generation satellite for orbital AI compute, named 'AI1,' which features a computing payload of up to 150 kilowatts and a design that allows for the replacement of compute providers. Following its merger with xAI, SpaceX is effectively transitioning into an AI cloud service provider, presenting a new investment rationale that vertically integrates rockets, satellites, and AI.
 
Colorado's AI Law Overhauled, Changing State-Level AI Regulation Landscape
 
Colorado's comprehensive AI consumer protection law has been effectively repealed and restructured. On May 14, the governor signed a new bill (SB26-189) that resets the implementation date to January 1, 2027. The new law abolishes the previous framework prohibiting algorithmic discrimination in 'high-risk AI systems' and redefines the regulation of automated decision-making technologies (ADMT) used in employment, healthcare, and finance as those that impact 'outcome-affecting decisions.' It also includes new provisions ensuring 'meaningful human review' for decisions unfavorable to consumers.
 
Analysts suggest that the lawsuit involving xAI and the federal court's stay decision expedited these amendments, signaling a shift in the leading model for state-level AI regulation in the U.S.
 
Model Competition Intensifies with Upcoming Releases of GPT-5.6 and Claude Sonnet 4.8
 
June is expected to be the most active month for model releases this year. Following the unveiling of Google Gemini 3.5 Flash at Google I/O, GPT-5.6 and Claude Sonnet 4.8 are set to launch simultaneously in mid-June. If both models are released around the same time, it could significantly reshape the token pricing competition in the agent-based production workload market. Attention is also focused on developments from China, as Alibaba's Qwen 3.7 Max has demonstrated performance on par with Claude Opus 4.7 in agent benchmarks, while offering a cost structure that is half the input price and one-quarter the output price, attracting interest from developers.



* This article has been translated by AI.

Copyright ⓒ Aju Press All rights reserved.