<aside>

Summary |

The different levels and kinds of AI Technologies, how intelligent they currently are, the risks, and what would enable them to improve even more /the limitations they face when it comes to advancement.

Aligning AI to goals requires interdisciplinary collaboration: both technical research and philosophical/public guidance because AI is a special type of technology that has a higher degree of freedom compared to things like a pencil or hammer. This autonomy requires a special approach that considers the risks and also invites a more sensitive attitude towards it. In order to produce technologies that are socially good then we must look at the values of humans and align them to the goals of developing AI.

</aside>

<aside>

Keywords -

• Ethics washing: Companies making superficial ethical claims without genuine action.

• deep learning

• Reinforced Learning (RL): SDG - Sustainable Development Goals

</aside>

<aside> ✔️

Study Questions:

1. What are the success conditions for doing the right thing? </aside>

Electronics watch

<aside>

Reading notes

The Uselessness of AI ethics

From Ethics washing to Ethics Bashing

</aside>

<aside> 📌

Introduction

Problems with seeing the ethics of computer science and AI as a set of facts and way of being

</aside>

<aside> 📌

The uselessness of AI Ethics

The gap between ethical principles and tech advancement is now large.

Ethical principles are meaningless, isolated, and toothless

Useless/meaningless in the sense that it does not influence the behaviour and motivations of the people developing the technology.

Meaningless

Ethical principles is so open to interpretation that it is materially impossible to arrive at something concrete.

It just adds a layer to processes, outputs, and outcomes.

output - a product

outcome - an effect of the outputs

the dangers of conflating outputs and outcomes

• principles are supposed to be outcomes but if principles become a goal then the point is lost

Goodheart’ Law: When a measure becomes a target, it ceases to be a good measure.

Principles can also be interpreted differently based on context, culture, individual, etc.

eg. Beneficence, non-maleficence, autonomy, justice, and explicability

We need to be able to specify the success conditions of the outcomes.hh

Isolated

• Ethical problems with tech industry do not occur in a vacuum. They are emergent.
• Downstream vs Upstream

Toothless

Not have traction

They just dont affect the behaviour of anyone working in the tech industry.

Ethical principles may replace regulations

Bias in AI is just a huge mechanism of distraction

• Micro issues, macro issues

Macro Issues

</aside>

<aside> 📌

Extractivism - a macro-issue of AI Ethics

• Physical Resources - rare earth minerals eg.
• Labour - Pruning datasets (modern labour)
• Content
• Identity: the extraction of diverse characteristics to fulfill an image rather than achieve an outcome
• Art: art vs decoration - art is deeper - it is created by AN artist and has a meaning. or rather meaning is created by a person based on the artist and genealogy of the artwork. The meaning of art is determined by the properties of it. There is no real success conditions of art.

</aside>

<aside> 📌

Ethics washing

Ethics washing can leaf to Ethics Bashing

Ethics washing is engaging in ethical Performance. Doing stuff to look like you’re ethical but you’re actually not

</aside>

<aside>

[answers] Study Questions:

You feel good, you achieved the goal you sought out to achieve, you don’t affect or at least only positively impact people around you.

</aside>