隐私政策
更新日期：2026年03月23日 | 生效日期：2026年03月23日
大连创锐光谱科技有限公司（以下简称“创锐光谱”或“我们”）高度重视用户的隐私和数据安全。作为 ChatTAS 瞬态吸收光谱智能数据分析系统（以下简称“本服务”、“本系统”或“ChatTAS”）的合法运营者，我们深知实验数据对科研人员的核心价值。本《隐私政策》严格依据《中华人民共和国个人信息保护法》、《数据安全法》等法律法规制定，旨在向您透明地展示我们在您使用 ChatTAS 期间，如何收集、处理、存储及保护您的信息。
1. 我们收集哪些信息及收集目的
为了向您提供基础的账号体系及核心的图表计算、智能问答服务，我们会按“最小必要”原则收集以下信息：
- 身份注册信息： 当您注册时，我们需要收集您的手机号码及密码，用于创建独立的工作区并满足网络实名制要求。密码通过高强度不可逆哈希算法（Bcrypt）加密存储，任何人无法触取您的明文密码。
- 业务处理数据：
- 您主动上传的 CSV 实验数据矩阵。
- 您与系统交互的自然语言对话记录（Chat History）、前端样式配置参数及操作指令。
- 日志与风控信息： 为保障系统稳定运行、限制恶意高频请求及异常登录排查，我们会在后台记录您的 IP 地址、浏览器 User-Agent、访问时间及资源消耗（Token）流水。
- 问题反馈信息： 若您主动提交故障反馈，我们将收集您的文字描述，以及您主动勾选授权的对话快照（含相关日志文件）。
2. 我们如何管理您的内容
作为面向科研领域的专业分析系统，我们深知数据保密性的极端重要性。我们将您在系统内产生的内容（包括实验数据文件与文本交互指令）进行分类，并实施严格且透明的管理机制：
- 实验数据文件：
您可在系统设置中自主决定是否授权您的数据参与“系统优化计划”。
- 若您选择授权参与： 我们会在严格脱敏的前提下，利用您的数据进行内部算法验证和功能迭代，以提供更优质的分析体验。
- 若您拒绝参与： 您的数据将仅用于响应您当前的处理与分析请求，我们绝不会将其用于任何系统优化或其他无关用途。
核心承诺： 无论您作何选择，为满足基本的服务运转与网络安全合规要求，您的原始数据文件仅会被加密暂存于我们的本地私有化服务器中，绝不会被上传或泄露给任何第三方机构。
- 文本指令与交互内容：
您输入的文本指令及系统生成的回复，将被本地化存储以便为您提供连贯的上下文解析服务（详情请见下方“流转说明”）。为了持续优化系统的意图识别准确率，您的非敏感指令文本可能会被用于系统的迭代升级。我们在使用前会通过技术手段进行严格的去标识化与去隐私化处理。 尽管如此，仍强烈建议您在输入自然语言指令时，避免包含任何个人身份信息或未公开的机密科研结论。
3. 数据在第三方大模型平台的流转说明（核心披露）
ChatTAS 集成了前沿的自然语言处理与意图识别能力。在提供 AI 问答及指令解析服务时，不可避免地需要将部分信息发送至经过国家网信办备案的第三方大模型服务提供商。在此我们做出庄重承诺及流转披露：
| 处理场景 |
涉及的具体数据 |
流向的第三方 |
流转声明 |
| 意图识别与参数提取 |
您的文本指令、系统的回复上下文 |
语言大模型API提供商或本地部署的大模型 |
加密传输。第三方服务商承诺不将 API 传入的信息用于模型训练。 |
| 智能问答 (RAG) |
您的提问文本 |
语言大模型API提供商 |
仅传输文本。不含任何您的身份标识等敏感信息。 |
| 光谱图表计算与拟合 |
您上传的原始 CSV 实验数据及矩阵 |
【数据不流转】仅在 ChatTAS 本地服务器运算 |
绝不将您的任何原始实验数据发送给外部大模型。所有算法计算均在我们的闭环服务器内完成。 |
4. 我们如何存储您的信息
- 存储地域： 我们收集的个人信息与实验数据均存储于中华人民共和国境内。
- 存储期限： 我们仅在为您提供 ChatTAS 服务期间保留您的个人信息。您的 CSV 文件和对话上下文会通过 Gzip 高比例压缩后暂存于缓存数据库中；当您主动触发“清空全部历史记录”或“注销账号”时，系统将立即且永久抹除相关数据文件。
5. 数据安全保护机制
- 传输与存储加密： 系统采用 TLS/SSL 协议保护数据传输链路，关键存储数据利用 Pickle+Gzip 进行二进制隔离压缩。
- 访问控制： 实施严格的服务器访问控制策略，隔离业务环境与公网环境。除保障服务器正常运行的运维人员外，任何人无权调阅您的私有工作区数据。
6. 您的权利
依据法律法规，您对您的数据拥有绝对的控制权：
- 查询与管理： 您可随时登录系统，在“历史对话”面板中查看或导出（PDF）您的工作流记录。
- 一键删除： 您可在系统界面的对话流中单步撤销、单独删除特定文件，或在个人中心使用“清空全部历史记录”彻底抹除记忆库。
- 账号注销： 您可以通过“个人设置页面 - 注销账号”按钮直接终止服务并清除所有关联数据。
7. 联系我们
如果您对本政策或数据隐私有任何疑问或投诉，请通过以下方式联系我们：
联系邮箱： sales@timetechspectra.com
一般情况下，我们将在验证您的身份后尽快予以回复并处理。
Privacy Policy
Updated: March 23, 2026 | Effective Date: March 23, 2026
Dalian Chuangrui Spectroscopy Technology Co., Ltd.
(hereinafter referred to as “Chuangrui Spectroscopy” or “we”)
highly values user privacy and data security.
As the lawful operator of the
ChatTAS Transient Absorption Spectroscopy Intelligent Data Analysis System
(hereinafter referred to as the “Service”, the “System”, or “ChatTAS”),
we fully understand the critical importance of experimental data to scientific researchers.
This Privacy Policy is formulated in strict accordance with the
Personal Information Protection Law of the People's Republic of China,
the Data Security Law,
and other applicable laws and regulations.
It is intended to transparently explain how we collect,
process,
store,
and protect your information during your use of ChatTAS.
1. Information We Collect and the Purpose of Collection
In order to provide you with a basic account system,
core chart computation features,
and intelligent Q&A services,
we collect the following information under the principle of
“minimum necessity”:
-
Account Registration Information:
When you register,
we collect your mobile phone number and password
to create an independent workspace
and comply with real-name registration requirements.
Passwords are encrypted and stored using a high-strength irreversible hashing algorithm (Bcrypt),
and no one can access your plaintext password.
-
Business Processing Data:
-
CSV experimental data matrices actively uploaded by you.
-
Your natural language conversation history (Chat History)
with the System,
frontend style configuration parameters,
and operational instructions.
-
Logs and Risk Control Information:
To ensure stable system operation,
limit malicious high-frequency requests,
and investigate abnormal login activities,
we record your IP address,
browser User-Agent,
access time,
and resource consumption (Token) logs on the backend.
-
Feedback Information:
If you voluntarily submit issue reports,
we will collect your textual descriptions
and any conversation snapshots
(including related log files)
that you explicitly authorize us to access.
2. How We Manage Your Content
As a professional analysis platform designed for scientific research,
we fully recognize the extreme importance of data confidentiality.
We classify the content generated within the System
(including experimental data files and text interaction instructions)
and implement strict and transparent management mechanisms.
-
Experimental Data Files:
You may independently decide in the System settings
whether to authorize your data to participate in the
“System Optimization Program”.
-
If you choose to participate:
We may use your data,
under strict de-identification procedures,
for internal algorithm validation and feature iteration
in order to improve analysis quality and user experience.
-
If you decline participation:
Your data will only be used
to respond to your current processing and analysis requests,
and will never be used for system optimization
or any unrelated purposes.
Core Commitment:
Regardless of your choice,
in order to maintain basic service operations
and comply with cybersecurity requirements,
your original data files will only be encrypted
and temporarily stored on our local private servers.
They will never be uploaded to or disclosed to any third-party institution.
-
Text Instructions and Interaction Content:
The text instructions you enter
and the responses generated by the System
are locally stored
to provide continuous contextual understanding services
(see “Data Flow Description” below for details).
In order to continuously improve intent recognition accuracy,
your non-sensitive instruction texts
may be used for iterative system optimization.
Before such use,
we apply strict de-identification and privacy-removal procedures through technical measures.
Nevertheless,
we strongly recommend that you
avoid including any personally identifiable information
or unpublished confidential research findings
in natural language instructions.
3. Data Flow to Third-Party Large Model Platforms (Core Disclosure)
ChatTAS integrates advanced natural language processing
and intent recognition capabilities.
In order to provide AI-powered Q&A
and instruction parsing services,
it is necessary to transmit certain information
to third-party large language model service providers
that have completed regulatory filing procedures
with the Cyberspace Administration of China.
We hereby make the following solemn commitments and disclosures:
| Processing Scenario |
Specific Data Involved |
Third-Party Recipient |
Data Flow Statement |
|
Intent Recognition and Parameter Extraction
|
Your text instructions
and the System’s conversational context.
|
Large language model API providers
or locally deployed large models.
|
Data is transmitted through encrypted channels.
Third-party service providers commit not to use API-submitted information for model training.
|
|
Intelligent Q&A (RAG)
|
Your question text.
|
Large language model API providers.
|
Only text information is transmitted.
No personally identifiable sensitive information is included.
|
|
Spectral Computation and Fitting
|
Your uploaded original CSV experimental data and matrices.
|
[No External Data Transfer]
Processed exclusively on ChatTAS local servers.
|
We never transmit any of your original experimental data
to external large language models.
All algorithmic computations are performed
entirely within our closed-loop server environment.
|
4. How We Store Your Information
-
Storage Location:
The personal information
and experimental data we collect
are stored within the territory of the People’s Republic of China.
-
Storage Duration:
We retain your personal information
only for the duration necessary
to provide ChatTAS services.
Your CSV files and conversational context
are temporarily stored in cache databases
after high-ratio Gzip compression.
When you actively trigger
“Clear All History”
or “Delete Account”,
the System will immediately
and permanently erase the relevant data files.
5. Data Security Protection Mechanisms
-
Transmission and Storage Encryption:
The System uses TLS/SSL protocols
to protect data transmission channels.
Critical stored data is compressed and isolated in binary form
using Pickle + Gzip mechanisms.
-
Access Control:
We implement strict server access control policies
to isolate operational environments from public network environments.
Except for authorized maintenance personnel
responsible for ensuring normal server operations,
no one is permitted to access your private workspace data.
6. Your Rights
In accordance with applicable laws and regulations,
you retain full control over your data:
-
Access and Management:
You may log in to the System at any time
to view or export (PDF)
your workflow records
through the “Conversation History” panel.
-
One-Click Deletion:
You may undo operations step-by-step,
individually delete specific files,
or completely erase the memory database
using the “Clear All History” function
in the Personal Center.
-
Account Deletion:
You may directly terminate the Service
and erase all associated data
through the
“Personal Settings → Delete Account”
function.
7. Contact Us
If you have any questions,
concerns,
or complaints regarding this Policy
or data privacy matters,
please contact us through the following method:
Contact Email:
sales@timetechspectra.com
Under normal circumstances,
we will respond and process your request
as soon as possible
after verifying your identity.