Overcrowding problem on China’s #ArtificialIslands in the #SouthChinaSea could be solved with tunnels, according to a #Chinese#ocean#engineering project team that says it has developed an excavation method to overcome the limitations of the soft coral sand.
After a rapid expansion of its presence in the contested #Spratly#reef chain that began about a decade ago, #China eased back on construction in recent years in a bid to soothe the nerves of neighbouring countries
So…the “Slack will now train AI using your data” thing is not as much of a five alarm fire as I’d first assumed:
“We do not develop LLMs or other generative models using customer data.” ← GOOD.
“Data will not leak across workspaces.” ← Or so they say. They •are• training across workspaces, but sounds like recommender systems and not generative models, so…we’ll see. Seems fraught. Still, that public commitment does mean something — legal exposure, at least.
The bad news: platform owners don’t have a vision and are out of ideas for how to innovate their #ecosystems, all in the name of #efficiency.
To paraphrase Jeff Goldblum’s character from Jurassic Park: The most brilliant minds of #engineering are optimized to execute and have not bothered to consider whether they should.
The good news: “real” #innovation will happen from the experienced middle, not the top.
Although companies have created detectors to help spot #deepfakes, studies have found that biases in the data used to train these tools can lead to certain demographic groups being unfairly targeted.
A team of researchers discovered new methods that improve both the fairness and the accuracy of these detection algorithms by teaching them about human diversity