•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Bitget is repositioning artificial intelligence from a standalone add-on to the exchange’s core operating layer, an approach Messari Research says supports the company’s push toward a “Universal Exchange (UEX)” model that combines spot, derivatives, on-chain access, and institutional-grade services.
The assessment is based on a report published April 28 UTC by Messari Research analyst Austin Freimuth. The report argues that Bitget is building an “AI-native” trading environment by unifying market intelligence, execution pathways, developer tooling, and user interfaces into a single AI stack—aiming to compress the full trade lifecycle, from analysis and strategy formation to risk checks and order routing, into one interface.
Founded in 2018 and registered in Seychelles, Bitget operates as a centralized exchange (CEX) offering spot and derivatives products, as well as real-world asset (RWA)-linked offerings, on-chain access features, and services aimed at institutional clients, according to the report.
The exchange emphasizes transparency through “proof of reserves (PoR)” disclosures and a separate protection fund. Its native token, Bitget Token (BGB), is positioned for fee discounts, campaign participation, and ecosystem utility.
Messari’s central claim is that Bitget’s AI effort is closer to a platform redesign than a chatbot overlay. Freimuth breaks the exchange’s AI stack into four layers—GetAgent, Agent Hub, GetClaw, and Gracy AI—each intended to serve different functions: decision support, developer infrastructure, autonomous execution, and strategic communication.
The report highlights a key distinction: information and execution are being re-coupled within the exchange environment, rather than requiring users to switch between third-party analytics tools and separate trading screens.
GetAgent is Bitget’s conversational trading interface on web and mobile. Messari says users can request market analysis, strategy ideas, and risk checks using natural language, with limited execution pathways integrated into the workflow.
Bitget states the feature is supported by more than 50 specialized trading tools, including trend and positioning data, high-activity token monitoring, meme-coin signal detection, and personalization based on a user’s assets and open positions. The goal, per the report, is to reduce interface switching—an ongoing friction point for non-professional traders even as algorithmic activity expands across crypto markets.
Messari cites early uptake figures, noting they are not independently verified. During an invite-based public preview in July–August 2025, GetAgent recorded more than 100 million impressions and more than 25,000 waitlist sign-ups. After that period, total users surpassed 450,000.
GetClaw is described as a Telegram-based consumer AI trading agent that monitors and executes strategies defined in natural language. Messari says the agent can track real-time prices, incorporate technical analysis, pull crypto news and on-chain signals, and detect conditional events such as funding-rate deviations, liquidation clusters, or sudden volume spikes—then trigger alerts or execute predefined workflows.
The report notes that because autonomous execution concentrates operational risk in “permissions” rather than model accuracy alone, Bitget has built guardrails around GetClaw’s ability to place trades. Trades are routed through dedicated sub-accounts to separate user-managed assets from agent-operated balances.
Messari also points to a four-way isolation framework—separating identity, memory, permissions, and credentials—along with funding limits and sandboxing intended to constrain access.
Bitget has also used AI as an engagement layer. In November 2025, the company introduced six AI trading “avatars” within GetAgent and ran a time-limited copy trading campaign. Messari says each avatar reflected a different trading persona and strategy profile, allowing users to select an AI trader aligned with their preferences and observe real-time autonomous trading behavior.
The campaign generated around 180,000 page visits, according to Messari. The report also says the avatar-style packaging—such as “Steady Hedge” or “Infinite Grid”—was positioned as a way to lower the learning curve for AI-assisted trading.
Gracy AI is presented as an interpretation and communication layer rather than an execution tool. Modeled on the public communications style of CEO Gracy Chen, it is designed to explain market conditions, macro narratives, sentiment shifts, and Bitget’s platform roadmap.
Messari reports that between Feb. 12 and Feb. 23, 2026 UTC, Gracy AI reached more than 460,000 users, generated over 2.6 million responses, and recorded 390 million impressions. The report characterizes the system as a “strategic interface” intended to provide context rather than place orders.
For developers, Agent Hub is described as the infrastructure layer that connects AI models to exchange functionality via Bitget’s API stack. Messari says external models and software agents can access market data, account information, and execution features through the hub.
The report says four access modes are supported: an MCP server, skills, REST and WebSocket APIs, and a command-line interface (CLI). Messari claims Bitget is the only major exchange currently offering all four, enabling developers to plug AI systems into products such as spot, futures, perpetuals, margin, copy trading, conditional orders, and balance management.
Messari frames the broader implication as competitive differentiation shifting from listings and liquidity toward “AI-based experience”—how efficiently a venue helps users translate information into action.
At the same time, the report cautions that long-term success will depend less on novelty and more on sustained behavioral change: whether users consistently trust AI interfaces to guide decisions and execute trades. It says execution reliability, the usefulness of outputs, and credible permission controls will determine whether early demand becomes durable adoption.
The report also flags open questions around regulatory expectations and potential misuse of automated trading tools.
Even with those constraints, Messari concludes that Bitget is among the most proactive exchanges in operationalizing “AI-native trading” as a platform architecture rather than a feature set. If crypto exchanges increasingly compete on integrated AI workflows, the firm suggests Bitget’s vertically integrated approach—spanning consumer UI, autonomous agents, and developer infrastructure—could serve as a reference point for the next phase of exchange design.
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
