China - Tier A · Elevated · Case: ZhongAn Online
Posted: Sun Oct 11, 2026 10:52 am
Fijishi Jurisdiction Index 2026 · Entry CN · Published 12 October 2026
At a glance
1. The rules in China
The National Financial Regulatory Administration's guidance on the safe development and application of AI in banking and insurance, Jin Fa [2026] No. 8, is dated 18 June 2026. It treats generative-AI applications involving underwriting and claims as high-risk and requires the institution's risk management committee to approve them before implementation (Item 16).
It also requires human supervision and intervention at the key links of high-risk applications (Item 17).
Who these rules reach
This band describes how clearly the rules meet the AI use that ZhongAn Online has itself published. It is not a finding that ZhongAn Online has breached any rule.
Why ZhongAn Online: ZhongAn's own 2025 results announcements describe Lingxi, its large-model AI platform, at the core of a business with AI built into underwriting and claims.
The question
3. The same question for other firms in China
Any insurance company in China using generative AI in underwriting or claims faces the same two points: whether those applications have its risk management committee's approval, and where the human checkpoint at their key links sits. This entry gives no band to any firm other than ZhongAn Online.
4. What this entry does not establish
Sources
None share a law or regulator with this entry.
________________________________________
Commercial relationship with any party named here: None · Right of reply: indexreport.protocol@fijishi.com; with the sender's agreement, replies are published in full below the entry, dated · Cite as: Fijishi Jurisdiction Index 2026, Entry CN · Corrections: none.
At a glance
| Jurisdiction tier | Tier A - Specified and enforced. A binding instrument that applies to insurers' own use of AI, or an issued regulator document addressed to this insurance market, addresses AI - by name or by an unambiguous description such as "predictive models" or "automated decision system" - and sets out rules or expectations that apply to it; and the regulator has a live mechanism to check or sanction them. |
| Case examined | ZhongAn Online - the Lingxi AI platform in underwriting and claims |
| Band for this case | Elevated |
| Rules reach | Insurers: directly · Reinsurers: not stated · Intermediaries: not stated |
| Last verified | [actual date of the launch-window check] |
The National Financial Regulatory Administration's guidance on the safe development and application of AI in banking and insurance, Jin Fa [2026] No. 8, is dated 18 June 2026. It treats generative-AI applications involving underwriting and claims as high-risk and requires the institution's risk management committee to approve them before implementation (Item 16).
It also requires human supervision and intervention at the key links of high-risk applications (Item 17).
Who these rules reach
- Insurers: directly. The guidance is addressed to, among others, "各保险集团（控股）公司、保险公司、保险资产管理公司" (insurance groups, insurance companies and insurance asset-management companies). Read on a secondary copy.
- Reinsurers: not stated. The addressee line does not name reinsurance companies separately.
- Intermediaries: not stated.
This band describes how clearly the rules meet the AI use that ZhongAn Online has itself published. It is not a finding that ZhongAn Online has breached any rule.
Why ZhongAn Online: ZhongAn's own 2025 results announcements describe Lingxi, its large-model AI platform, at the core of a business with AI built into underwriting and claims.
The question
In briefLingxi - the AI platform ZhongAn built on mainstream domestic large models - is, in ZhongAn's 2025 annual results (19 March 2026), the core engine of a business with AI built into underwriting and claims. Item 16 of NFRA's Jin Fa [2026] No. 8 treats generative-AI applications involving underwriting and claims as high-risk and requires the institution's risk management committee to approve them before implementation; Item 17 requires human supervision and intervention at the key links of high-risk applications. Can ZhongAn confirm that Lingxi's underwriting and claims applications have received that committee approval, and say where the Item 17 human checkpoint sits in them?
Why this band: The question points to one firm, one platform and two named duties, but the guidance's effective-date clause was not located, so the date from which it binds is less firm.ZhongAn's 2025 results put Lingxi, its large-model AI platform, at the centre of a business with AI built into underwriting and claims, and NFRA's June 2026 guidance treats generative-AI underwriting and claims applications as high-risk, needing risk-management-committee approval and a human checkpoint at key links. Neither the approval nor the checkpoint for Lingxi appears in anything ZhongAn has published.
3. The same question for other firms in China
Any insurance company in China using generative AI in underwriting or claims faces the same two points: whether those applications have its risk management committee's approval, and where the human checkpoint at their key links sits. This entry gives no band to any firm other than ZhongAn Online.
4. What this entry does not establish
- The guidance was read on government-hosted and other secondary copies, not on the regulator's own site; its effective-date clause was not located.
- ZhongAn's 2025 annual results were read on a reproduction of the company's announcement; the stock-exchange original was not read.
- That Lingxi's underwriting use involves generative AI is read from ZhongAn's description of it as built on large models.
- Whether the committee approval or the human checkpoint exists. That is the question; this entry says nothing either way.
- No date set: confirmation of the guidance's effective date.
Sources
- ZhongAn Online, 2025 interim results, Hong Kong, Aug. 20, 2025 (printed). Company's own. PR Newswire
- ZhongAn Online, 2025 annual results, 19 March 2026 (date printed in text), read on a reproduction. Company's own. xhby.net
- National Financial Regulatory Administration, 关于银行业保险业人工智能安全开发应用的指导意见, Jin Fa [2026] No. 8, 18 June 2026. Regulator, Chinese. policy.mofcom.gov.cn
None share a law or regulator with this entry.
________________________________________
Commercial relationship with any party named here: None · Right of reply: indexreport.protocol@fijishi.com; with the sender's agreement, replies are published in full below the entry, dated · Cite as: Fijishi Jurisdiction Index 2026, Entry CN · Corrections: none.