AI and Ethics:
 

@BenByford

safe guarding & wellness

@BenByford

AI and Ethics:
 

safe guarding & wellness

What is Ai?

• A social construct?

• A Dream

• A Human like robot?

• A set of technologies?

• algorithms and data?

• Anything doing something human LIKE?

• A social construct?

• A set of technologies?

What is Ethics?

wikipedia

Ethics or moral philosophy is a branch of philosophy that involves systematizing, defending, and recommending concepts of right and wrong conduct.

Ethics is about US

• What is right to do?
• What behaviour is and isn't permissible or desirable?
• What kinds of opportunities are we making
  available or not?
• What direction are we taking the world in?
• What does human flourishing look like?
• What ought we create to achieve this?

how are ethics codified?

• Laws

• Money / Grants / Tax

• Standards / frameworks / code of conduct

• Culture / cultural artefacts / traditions

• Bureaucracy / Systems / Computers

In what way does
AI participate in ethical questions?

"we shape the tools, then the tools
shape us”

Culkin, 1967

“It is impossible to work in information technology without also engaging in social engineering.”

Jaron Lanier

Gen AIssues

Capabilities

Often I pop to the garage for tea,
but dessert was a tomato.

So what's going on here?

AIssues

Capability

Machine Learning

Accuracy, job erosion, capability to further surveillance society, lack of accountability, fairness / bias, unintended consequences, lack of transparency / explainability, data protection, consent, services ethical values, information obfuscation, duplication of personal data security, safety, robustness, mental and physical harms, Erosion of trust in information

Generative AI

Accuracy, AI value alignment, impersonation / deepfakes, misinformation, hallucinations, essay cheating, environmental impacts, echo chambers, plagiarism, intellectual property rights, perversion, knowledge / skill erosion, tool for malicious use, personhood, access to harmful information, hacking attacks, reinforcing social biases, overreliance on tool

the content produced by generative AI can be:

• inaccurate
• inappropriate or unsafe
• biased
• taken out of context
• taken without permission (intellectual property infringement)
• out of date or unreliable
• low quality

https://www.gov.uk/government/publications/generative-artificial-intelligence-in-education/generative-artificial-intelligence-ai-in-education

the content produced by generative AI Can be:

• used to empower people to use the tools for milicious purpose

  • cyber-attacks, fraud, scams, impersonation, child sexual abuse images
  • images / video / text misinformation and harmful content

https://www.gov.uk/government/publications/frontier-ai-capabilities-and-risks-discussion-paper/safety-and-security-risks-of-generative-artificial-intelligence-to-2025-annex-b

Mixed messages

• AI Opportunities Action Plan 2025

"Invest in the foundations of AI;
Push hard on cross-economy AI adoption;
Position the UK to be an AI maker,
not an AI taker"

Mixed messages

• Generative artificial intelligence (AI) in education

“The intended use should be specified and have clear benefits that outweigh the risks.”

When is IT both useful and
appropriate To use AI?

Thank you

Q&A

www.benbyford.com