Google launches Gemini 3.5 Flash AI model

Key Takeaways

• Google launched Gemini 3.5 Flash, a fast and efficient AI model that outperforms its predecessors.
• Gemini 3.5 Flash scores 55 on Artificial Analysis's benchmark for real-world tasks, beating other models like Claude Sonnet 4.6 and Claude Opus 4.6.
• The model delivers 280 output tokens per second and is priced at $1.50 per million input tokens and $9 per million output tokens.
• Meta is reorganizing its workers' jobs around AI, with a focus on AI agents and cloud infrastructure.
• Moore Threads Technology Co., Ltd. launched AICUBE, a unified home AI hub.
• The use of AI is expanding into various industries, including education, clinical trials, and the mortgage industry.
• There are concerns about bias, accountability, and addiction in the use of AI.
• Google CEO Sundar Pichai joked about 'tokenmaxxing' and highlighted the importance of AI token usage.
• The integration of AI in education requires changes to prepare young minds for a new world.
• Practical AI governance is needed in the mortgage industry.

Google Launches Fast AI Model Gemini 3.5 Flash

Google has released a new AI model called Gemini 3.5 Flash. It is designed to be fast and efficient, allowing developers to build complex AI models. Gemini 3.5 Flash can handle more tasks and workflows than its predecessors. Google claims it outperforms its last-gen Pro model. The model will be available across various Google products.

Gemini 3.5 Flash Outperforms AI Models

Google's Gemini 3.5 Flash model scores 55 on Artificial Analysis's benchmark for real-world tasks. It beats other models like Claude Sonnet 4.6 and Claude Opus 4.6. Gemini 3.5 Flash is fast, delivering 280 output tokens per second. It is priced at $1.50 per million input tokens and $9 per million output tokens.

Google Releases Gemini 3.5 Flash Model

Google DeepMind announced Gemini 3.5 Flash at Google I/O 2026. It outperforms Gemini 3.1 Pro on coding, real-world tasks, and scaled tool use benchmarks. Gemini 3.5 Flash can plan and reason across massive codebases. It delivers 289 tokens per second, four times faster than other models.

AI in Courtrooms: A New Era

The use of AI in courtrooms is changing the way justice is delivered. AI tools are being used to analyze data and make decisions. However, there are concerns about bias and accountability. A recent event by Global Academy for Future Governance discussed the intersection of AI, litigation, and international arbitration.

Moore Threads Unveils AI Hub for Smart Homes

Moore Threads Technology Co., Ltd. launched AICUBE, a unified home AI hub. AICUBE integrates AI computing, storage, and display capabilities. It will be available in 2026 and aims to make AI more accessible for everyone.

Google CEO Jokes About AI Token Usage

Google CEO Sundar Pichai announced a 7x increase in Google's AI products at the I/O developer conference. The increase drew audible gasps from the audience. Pichai joked about 'tokenmaxxing,' but said the stat tells an important story.

TaskGround: AI for Household Tasks

TaskGround is a new AI system that enables compact models to interpret complex scenes and act effectively. It uses a Ground-Infer-Execute framework to distill information and enable more focused AI reasoning. TaskGround achieved significant improvements in task success rates across various models.

Education and AI for Young Minds

The integration of AI in education requires changes to prepare young minds for a new world. A course on futurology is proposed to teach students to apply foundational skills to a future of knowns and unknowns created by AI. This course would examine the best curricula of the past, present, and future.

Clinical Trials and AI Flaws

A new analysis found that clinical trial flaws are being scaled, not solved, by AI. The use of AI in clinical trials is primarily focused on automating routine tasks. There is a need for fundamental changes in clinical trial design and execution.

Mortgage Industry Needs Practical AI Governance

The mortgage industry needs practical AI governance, not just AI ambition. AI is already embedded in the mortgage manufacturing process. Practical AI governance means creating a framework for responsible AI development and deployment.

Meta Reorganizes Workers Around AI

Meta is reorganizing its workers' jobs around AI. Some employees will be moved to new teams focused on AI agents and cloud infrastructure. The company is rapidly changing its work structure to focus on AI.

Addictive AI: The Next Big Business Risk

AI is making digital products more engaging, but this also creates a risk around addiction and mental health. Businesses need to consider the risks of AI-driven addiction and take steps to mitigate them.

Sources

• Gemini 3.5 Flash might be fast enough for gen AI to make sense
• Gemini 3.5 Flash Places Fifth On Artificial Analysis Intelligence Index, Is Faster Than Other Models
• Google Releases Gemini 3.5 Flash, Beats Gemini 3.1 Pro On Many Benchmarks
• Navigating Courtrooms In The Era Of AI: Towards Equitable International Litigation And Arbitration
• Moore Threads Launches AICUBE: A Unified Home AI Hub
• It's official: Google is 'tokenmaxxing'
• TaskGround: Bridging Scene Context and Action
• What education must do with AI for the sake of young minds
• Clinical trial flaws ‘being scaled, not solved’ by AI: report
• The Mortgage Industry Needs Practical AI Governance, Not Just AI Ambition
• Meta is rapidly reorganizing its workers’ jobs around AI: ‘Transfers aren’t optional’
• Addictive AI Could Become The Next Big Business Risk