Note: This is more of an attempt to distinguish between ethical and decision consequentialism, both of which was outlined in the original post, so please feel free to share your thoughts or let me know if there is anything in this post that is significantly misrepresented. 

Ethical Consequentialism:

• Ethical consequentialism determines the morality of actions solely on their consequences - if the outcome of the action is good, the action is considered morally right and vice versa
• There is a large focus on the results and consequences rather than the actions themselves or intentions
• The morality of actions can be determined based on if it produces the most good (or least harm) compared to the other possible actions
• The most well-known type of ethical consequentialism is utilitarianism - aiming to maximize the overall happiness or well-being for the greatest number of people
• An example is if you have a choice to donate money to a charity. According to consequentialism, the decision to donate would be right if it leads to positive outcomes, like helping people in need and improving their lives.

Decision Consequentialism

• Decision consequentialism is focused on the decision-making process and the expected outcomes
• An agent weighs up and evaluates the potential results/outcomes of choices and picks the one with the best expected utility.
• Thus, the morality of decisions is determined based on the likely consequences of the decisions
• It encourages rational choices to yield the best overall consequences.
• Applicable/ used in AI alignment to describe the likely consequences of advanced artificial intelligence during decision making processes/ generally when it makes choices
• An example is deciding between investing in renewable energy or fossil fuels by evaluating which option will result in better outcomes for the environment and society - the choice is what is being evaluated and the likely consequences of it