liveBook · Manning

Risk management

As explained in Starting Data Analytics with Generative AI and Python

Risk management is the process of identifying, assessing, and addressing potential threats to objectives or operations.

Risk management identifies, assesses, and addresses threats to objectives or operations. In generative-AI analytics, it keeps potential gains in accuracy and efficiency aligned with safe, responsible, and ethical use. section 8.0

Treat an AI system like a new mind on the team: useful, but requiring oversight. Effective practice identifies risks before they materialize, evaluates their seriousness, and applies mitigations rather than accepting every output or rejecting the model entirely. section 8.0 section 8.1

Practical risk management emphasizes:

  • Recognizing that safeguards against harmful content can have limited success because developer assumptions may introduce new weaknesses and determined users may bypass protections. section 8.1
  • Balancing the extremes of supplying a model with all available data and passing its output directly to stakeholders, or abandoning the model and doing everything manually. section 8.0
  • Using expert review, adversarial testing, and observed user interactions to identify, analyze, evaluate, and mitigate risks. section 8.1

Monitoring negative mentions

A company can apply risk management by monitoring mentions of its name alongside negative terms. Rising signals can prompt investigation and an earlier response to potential issues. section 5.5.1

How Starting Data Analytics with Generative AI and Python covers it

How deep it goes
Explains identification, analysis, evaluation, and mitigation, while stressing that safeguards have limited and uneven success.
Examples
Uses AI safety reviews, harmful-content safeguards, and monitoring company mentions alongside negative terms as concrete examples.
What you can do after
Evaluate generative-AI risks, recognize mitigation tradeoffs, and respond proactively to emerging business issues.
How it connects
Connects risk management with generative AI, data analytics, ethics, cybersecurity, adversarial testing, and keyword search.

Also explained in

Compare how 6 books explain risk management →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Starting Data Analytics with Generative AI and Python. The section links point to the parts of the book it draws on.