weekend ai reads for 2024-10-18

The researchers estimate that between 0.5 and 3.5 percent of all work hours in the U.S. are currently supported by generative AI.

GS is a major financial institution, not an e/acc enthusiast on social media. It offers a relatively cautious but reasonable forecast that, in effect, assumes a future AI productivity impact comparable to the PC + internet combo. This makes for a sensible but hardly unexciting baseline.

Therefore, if you are OpenAI, Anthropic, or another AI vendor, you have two choices. Your first is to spend enormous amounts of money to stay ahead of the market. This seems very risky though: The costs of building those models will likely keep going up; your smartest employees might leave; you probably don’t want to stake your business on always being the first company to find the next breakthrough. Technological expertise is rarely an enduring moat.

Your second choice is…I don’t know? Try really really hard at the first choice?

There are definitely ways to optimize my monthly payment to save a bit of cash, but I’m currently paying roughly $157/month.

appreciate everything endlessly

One big change is that AI will enable individuals, or very small groups, to run large projects. By directing AIs, they will be able to create entire think tanks, research centers or businesses. The productivity of small groups of people who are very good at directing AIs will go up by an order of magnitude.

Philanthropists ought to consider giving more support to such people.

I’m not saying that AI summaries are that vague. In fact, the reason they’re dangerous is that they’re good enough. They allow you to fake it, to proceed with some understanding of the subject. Just not a deep one.

Manually, it would take me a day to figure this out and write the code.

I wrote a proper feature requirements documenting everything.

I gave this to o1-mini, it thought for ~120 seconds.

I think most people probably don’t use the models enough yet. I know some people have concerns about “slop” — but the short answer is that I don’t believe Claude is reducing the quality of my work

Here’s how to implement AI in business:

1. Familiarize yourself with the capabilities and limitations of artificial intelligence

2. Identify your goals for implementing AI

3. Assess your company’s AI readiness

4. Integrate AI into select tasks and processes within your organization

5. Learn from your mistakes and aim for AI excellence

What we found is developing it internally lets us do a few things: It lets us really focus on creating an AI bot that can respond in ways that are very customized to our teachers and staff.

The new system has raised ethical questions. Is it fair, for example, for a girl with the same academic and behavioral troubles as a male classmate to be classified at lower risk, simply because girls overall tend to have better outcomes than boys?

The CTO survey found that the most popular applications of AI after chat bots included enhancements of cybersecurity (35 percent), predictive analytics for student performance and trends (27 percent), research and data analysis (24 percent), and learning management systems (20 percent).

I know that it is a pretty long blogpost, so that is why I am adding this part:

- More data != always better.

- Data pruning can break neural scaling laws from power laws to exponential scaling.

- The more you refine a dataset, the less you have to train the model.

- As a community we should try to do “foundational datasets”. What this means is that heavily refined, high-quality datasets can amortize the cost of AI training over time.

TikTok determined the precise amount of viewing it takes for someone to form a habit: 260 videos. After that, according to state investigators, a user “is likely to become addicted to the platform.”