Great article Alex. Particularly enjoyed "Anyone who’s asked an AI to make a big list of links knows this problem well, sometimes rather than just say “I don’t know” the machine puts down what it ‘thinks it knows’ in an effort to be helpful...Which is precisely the behavior that gets (sometimes literally) beaten out of you as a first year analyst or whatever working in finance." 💯
Coming from the SV world, this was a great read and helped me crystallize my own intuitions. Another SV/NY difference in terms of product building (and one which may partially explain this focus on workflows) is that SV wants to find opportunities that scale with increasing margins. The data products (as opposed to workflow focused ones) throw a wrench in the wheel because while there may be more incremental value to be captured by offering it to more people, it will also dilute the alpha and the value of the dataset for existing users. You may as well charge a higher amount with fewer subscribers and they'll pay for it heartily since they'll all make more money off it. Curious to hear your thoughts on that.
Amazing connection of the dots .. One question, why does Bloomberg not do this itself .. build its own AI with its cleaned data and offer it for $20/$50/$100 around the world?
Great article Alex. Particularly enjoyed "Anyone who’s asked an AI to make a big list of links knows this problem well, sometimes rather than just say “I don’t know” the machine puts down what it ‘thinks it knows’ in an effort to be helpful...Which is precisely the behavior that gets (sometimes literally) beaten out of you as a first year analyst or whatever working in finance." 💯
Are you saying that AI’s hallucination problem (in the financial domain) will be solved by specialist human beings feeding it curated information?
That's the first step.
The 'foundational model' for finance.
Coming from the SV world, this was a great read and helped me crystallize my own intuitions. Another SV/NY difference in terms of product building (and one which may partially explain this focus on workflows) is that SV wants to find opportunities that scale with increasing margins. The data products (as opposed to workflow focused ones) throw a wrench in the wheel because while there may be more incremental value to be captured by offering it to more people, it will also dilute the alpha and the value of the dataset for existing users. You may as well charge a higher amount with fewer subscribers and they'll pay for it heartily since they'll all make more money off it. Curious to hear your thoughts on that.
Amazing connection of the dots .. One question, why does Bloomberg not do this itself .. build its own AI with its cleaned data and offer it for $20/$50/$100 around the world?
Maybe b/c Bloomberg is a bit of a luxury product. The network/Terminal chat specifically.