Approach

At TOAQ, safety is a foundational concern throughout the design, development, and evaluation of our research activities, technologies, and digital systems.
This Safety Approach outlines the principles and measures we apply to reduce risks, prevent misuse, and promote responsible experimentation.

1 – Scope of Safety

TOAQ’s safety approach applies to:

• Research activities and experimental systems;
• Software tools, prototypes, and demonstrations;
• Automated or AI-assisted processes hosted or referenced on this Site.

Given the experimental nature of certain components, safety is treated as an evolving process rather than a fixed guarantee.

2 – Risk Awareness and Mitigation

We actively assess potential risks associated with our technologies, including:

• Technical failures or unexpected behaviors;
• Misinterpretation or misuse of outputs;
• Ethical, social, or security-related concerns.

Where risks are identified, reasonable mitigation measures are implemented, such as:

• Limiting system capabilities;
• Applying usage constraints;
• Clearly communicating limitations and intended use.

3 – Human Oversight

Human supervision remains central to our safety strategy.

• Automated systems are monitored and reviewed where feasible;
• No system is intended to operate autonomously in safety-critical contexts;
• Decisions with significant impact should not rely solely on automated outputs.

4 – Responsible Use of Artificial Intelligence

When artificial intelligence is used or studied, TOAQ follows responsible AI principles:

• AI outputs are probabilistic and may contain errors;
• Systems do not possess awareness, intent, or judgment;
• AI-generated content should not be interpreted as authoritative or professional advice.

Users are encouraged to verify outputs independently and apply critical judgment.

5 – Data Protection and Security

Safety includes the protection of data and infrastructure.

• Personal data protection is addressed in accordance with GDPR requirements;
• Access to systems and data is restricted where appropriate;
• Reasonable technical and organizational safeguards are applied to reduce security risks.

Details regarding personal data are provided in the Privacy Policy:
Privacy Policy

6 – Prevention of Misuse

TOAQ does not intend its technologies or research outputs to be used for:

• Harmful, illegal, or malicious activities;
• Surveillance, exploitation, or manipulation of individuals;
• Safety-critical or life-critical applications without proper validation.

Where misuse risks are identified, access may be limited or features removed.

7 – Transparency and Accountability

Safety practices are closely linked to transparency and accountability.

• Known limitations are communicated openly;
• Experimental status is clearly indicated;
• Feedback and concerns are taken seriously and reviewed when appropriate.

8 – Continuous Improvement

Safety is not static.
TOAQ continuously reviews its methods, tools, and assumptions in light of new research findings, user feedback, and evolving technical or regulatory standards.

9 – Reporting Safety Concerns

Users, researchers, or partners who identify potential safety issues are encouraged to report them.

Safety-related concerns may be sent to:
[email protected]

10 – Updates to This Safety Approach

This Safety Approach may be updated as the project evolves.
Any changes will be published on this page and take effect immediately upon publication.