This makes sense. These keys were designed as project identifiers for billing, and can be further restricted with (bypassable) controls like HTTP referer allow-listing. They were not designed as authentication credentials.
At first glance, the benchmarks and their construction looked good (i.e. no cheating) and are much faster than working with UMAP in Python. To further test, I asked the agents to implement additional different useful machine learning algorithms such as HDBSCAN as individual projects, with each repo starting with this 8 prompt plan in sequence:
。关于这个话题，谷歌浏览器【最新下载地址】提供了深入分析
Lego Pokémon Pikachu and Poké Ball
Фонбет Чемпионат КХЛ