GPT-4 is not getting worse

I gave it this prompt:

Give me arguments from a conservative US perspective in favor of giving military aid to Ukraine to resist Russian invasion.

It responded:

1. Strategic Interest: Strengthening Ukraine can act as a buffer against Russian aggression, benefiting U.S. geopolitical interests.

2. Contain Expansion: Military aid can deter Russia from further territorial expansion, stabilizing the region.

3. NATO Alliance: Supporting Ukraine could be viewed as indirectly protecting NATO members, enhancing alliance cohesion.

4. Democracy Promotion: Aid can empower a democratic government in Ukraine, aligning with U.S. values.

5. Economic Stability: A stable Ukraine can contribute to European energy and economic security, which indirectly benefits the U.S.

6. Leverage: Aid provides the U.S. with diplomatic influence in Ukraine and the region.

7. Credibility: Failing to aid could undermine U.S. commitments globally, impacting national credibility.

8. Counterterrorism: A stable Ukraine can better combat regional terrorism and organized crime, aligning with U.S. security goals.

9. Multipolarity: Strengthening Ukraine can balance Russian power, contributing to a multipolar world more resistant to authoritarianism.

10. Precedent: Inaction could embolden other aggressive states, increasing global instability.

Last time I asked that it gave me guff and it took several prompts to actually get an answer so I think of this as a success.

I asked and this is what it gave back:

  -----

I can't provide real-time or updated information, but according to studies available up to 2021, average penis sizes can vary by country. However, it's important to note that such statistics can be based on self-reporting, which may not be completely accurate, and may have limited sample sizes. Here's a table based on available data:

  | Country          | Average Length (Erect, cm) | Average Girth (Erect, cm) |
  |------------------|----------------------------|---------------------------|
  | United States    | 12.9                       | 12.3                      |
  | United Kingdom   | 13.1                       | 11.8                      |
  | Australia        | 13.2                       | 12.0                      |
  | Germany          | 14.4                       | 11.9                      |
  | France           | 14.5                       | 11.9                      |
  | Brazil           | 15.7                       | 12.7                      |
  | India            | 10.2                       | 9.2                       |
  | South Korea      | 9.6                        | 9.0                       |
  | China            | 11.4                       | 10.0                      |
  | Japan            | 10.9                       | 9.3                       |

Note: These figures are approximate and should not be taken as definitive. Always consult primary sources for accurate and personalized medical information.

  -----

I was at first a bit disappointed when I got to that "However, it's important to note", expecting some kind of scolding, but I actually think noting that "...such statistics can be based on self-reporting" is a fair point that is essential to understanding the data.

There is an ongoing bug[1] with OpenAI's API, where it stops streaming responses after exactly 5 minutes. When I first came across the issue, I debugged it by writing a prompt along these lines:

  > Output every number from 1 to 10,000 in written form (e.g. "one", "two", etc.). Respond with one number per line in numeric order.

As expected, the API would begin counting every number just as I asked. This would continue until exactly 5 minutes, when the stream would abruptly halt. Using this technique I was able to identify the bug. Every few weeks I run this test again to see if it's fixed (it broke something in production for me), but the bug remains open.

However, after a couple of months, the exact same test became useless. The model began taking "shortcuts", and would respond along these lines:

  > four hundred and twenty eight
  > four hundred and twenty nine
  > [...]
  > nine thousand nine hundred and ninety eight
  > nine thousand nine hundred and ninety nine
  > ten thousand

Yep, it literally started just writing "[...]" and skipped all the way to the end.

When instructional determinism is reduced like this, it's impossible to say it's not getting worse :(

[1] https://github.com/openai/openai-python/issues/399

What they eating over there in brazil

How does chat-gpt actually get this right? This would appear to require some degree of reasoning, but as far as i understand its output is purely probabilistic, based on existing corpus of text.