The way we measure ocean temperatures changed. The oceans are actually within normal bounds if you use previous measurement methods. The new approach however shows more volatility that is not reflective of what is actually going on.
Do you have any concrete particular issue with, say, the European sea surface temperature ERA5 1979-2024 dataset ?
https://science.nasa.gov/earth/climate-change/the-raw-truth-...
There has been thoughtful and careful attention made to normalize the temperatures across any number of data collection and aggregation methods and they’re all showing the same upward trend.