Futures
Access hundreds of perpetual contracts
CFD
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
CFD
U.S. stock CFD derivatives
US Stocks
Access real US stocks and ETFs
HK Stocks
Trade quality Hong Kong-listed stocks
Korean Stocks
SK Hynix
Real Korean stocks and top assets
Stock Futures
High leverage, 24/7 trading
Tokenized Stocks
Backed by real stock assets
IPO Access
Unlock full access to global stock IPOs
GUSD
Mint GUSD for Treasury RWA yields
Stocks Activities
Trade Popular Stocks and Unlock Generous Airdrops
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
IPO Access
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Promotions
AI
Gate AI
Your all-in-one conversational AI partner
Gate AI Bot
Use Gate AI directly in your social App
GateClaw
Gate Blue Lobster, ready to go
Gate for AI Agent
AI infrastructure, Gate MCP, Skills, and CLI
Gate Skills Hub
10K+ Skills
From office tasks to trading, the all-in-one skill hub makes AI even more useful.
Zuckerberg explains why he monitors employees’ keyboard and mouse: outsourcing is too incompetent, while Meta employees are very smart
AIMPACT News, May 20 (UTC+8): According to Dynamic Insight Beating monitoring, U.S. labor outlet More Perfect Union exposed a recording of a Meta internal all-hands meeting. In the company-wide meeting on April 30, CEO Mark Zuckerberg admitted that the company has implemented an internal tracking project called the “Model Capability Initiative” (MCI). Through monitoring tools running on employees’ computers, it records their mouse movement trajectories, click locations, keyboard keystrokes, and screen screenshots.
Facing internal pushback, Zuckerberg vigorously defended the plan in the recording. He said that, currently, training data for industry large models is highly dependent on outsourcing, while Meta employees’ average intelligence is far higher than that of ordinary outsourced workers. In his view, the daily traces of thousands of elite engineers solving programming tasks and building tools are extremely scarce, high-quality training materials. Meta is trying to use this to completely outpace competitors in the speed at which its models’ code evolves.
Besides writing code, the system must also learn to “use a computer” like a human. To enable AI agents to master a chain of operations such as navigating drop-down menus and using shortcut keys, the most direct approach is to have them continuously observe how smart people use their computers.
When employees questioned whether they were being secretly monitored, Zuckerberg assured them that no human would look at these records. Sensitive content would be stripped as much as possible before any data leaked, and the data would never be used for performance evaluations or employee monitoring.
As Meta is facing layoffs, many employees are strongly resistant to being forced to hand over their personal computer data—or even use it to train “AI that can replace them.” Zuckerberg acknowledged in the meeting that early communication was not sufficient, but emphasized that the project must remain confidential: AI competition is too brutal. Once this exclusive tactic that can create a lead is made public, rivals will immediately copy it.
And this is only the beginning. Zuckerberg said bluntly that as long as this tactic is proven to improve model capabilities, Meta will then roll it out across the entire company, turning high-level employees’ computer operation traces into training materials for AI models.
(Source: BlockBeats)