🐘 Natural language processing thinks we are doing a better job at conservation and the coronavirus-inspired squeeze on privacy is here
🤖Stupid robots aren't taking the jobs we need them for rn
The Good
Good news for our non-AI using brethren! Researchers at the Monterey Bay Aquarium looked at 4,000 studies looking at conservation efforts from 1987 to 2016 and found that we're getting better at reintroducing species to the wild. Instead of reading through a million words and sorting this out on their own, the researchers relied on sentiment analysis and five different natural language processing models to detect positive and negative terms across the papers to get a sense of how well the work was going.
The words they were looking for that suggested reintroduction was going well included "success," "protect," "growth," "support," "help," and "benefit." Negative terms were ones like "threaten," "loss," "risk," "threat," "problem," and "kill.”
Here’s a chart of what they found, with a higher sentiment score coming from the increased occurrence of positive words (although it is unclear what happened around 1996?):
They also did some breakdowns for specific ecosystem and conservation challenges, which you can see in the paper. Some uses of sentiment analysis are bound to pick up a lot of noise (Reddit, Twitter) and can’t detect sarcasm, making some uses of it worthless (even though researchers and Mark Zuckerberg try). But conservationists, who aren’t known for sarcasm OR dank meme jokes, create a more straightforward dataset to analyze.
The Bad
Singapore just open-sourced its surveillance tech that lets people know if they got close to someone that had coronavirus. Russia is using the pandemic as a reason to expand their face rec tech-powered camera system. And it’s not just Border Patrol trying to pitch face rec tech as a hygienic solution to identify people.
All these surveillance responses to coronavirus have privacy folks worried. In “To Pre-Empt Future Pandemics, Governments Should Invest in the Welfare State, Not Private Surveillance” pubbed by the Berkman Klein Center, Elettra Bietti writes:
Instead of unreflectively welcoming private companies into our lives, we should seize this moment to reflect on the future of our increasingly digital democracies, of our hollowed-out social security systems, and of our fragmented global economy. If we don’t, we risk the erosion of democratic values and the normalization of surveillance to a point of no return.
It’s a trick from an old playbook. In The Shock Doctrine, Naomi Klein describes the way private industries spring up to profit from large-scale crises. While most of the book covers coups and hurricane-type events leading to disaster capitalism, she points out in this 2017 article that Trump has an awful lot of people from post 9/11 surveillance companies in his administration. They are already doing damage to our institutions, and she warns what they could do if something big (like now) happens:
The Republicans under Donald Trump are already seizing the atmosphere of constant crisis that surrounds this presidency to push through as many unpopular, pro-corporate policies. And we know they would move much further and faster given an even bigger external shock.
If we aren’t diligent, we are going to end up with a surveillance network first with a haphazard legal framework added on later, instead of the other way around.
More News
Could really use better robots to combat coronavirus but their imminent job takeover has been greatly exaggerated.
COVID-19 updates:
Supercomputers are helping crunch some data. The European Commission wants your AI solutions. A new consortium plans to give grants and for some reason Condoleezza Rice is involved. Xprize launches Pandemic Alliance to speed up research. Autonomous vehicles might get a boost from our social distancing needs. Our ISPs are using AI to help level out our WFH internet needs.
Computer vision trained to spot moon craters (useful for creating paths for rovers) can also spot where landmines might still be buried from the Vietnam War.
Aww! Animal shelters are using special face rec tech to match missing pets with their owners.
Drones can autonomously dodge shit you throw at them now.
All the reporting on AI beating doctors has some ~issues~, mainly small datasets that the researchers put together themselves.
Starsky Robotics, a major autonomous vehicle company, just shut down. The CEO said:
“Supervised machine learning doesn’t live up to the hype. It isn’t actual artificial intelligence akin to C-3PO, it’s a sophisticated pattern-matching tool.”
***
Until we figure out how AI can help save cheetahs so I get to see my favorite animal in the wild,
Jackie