What is privacy in AI? The reason there is no simple answer is not so simple to explain. Even without AI, there are degrees of privacy. In the most privacy friendly setting, you run an app such as PianoMeter. It does some calculation on your phone, shows you the results and that’s it. A less private case is something like Trello: you send your data to the company, they are processed somewhere, and you get to see the result. In the ideal case nobody else can access your data, but in reality there is a long slide, beginning with company insiders being able to look up your data (think Uber employee stalkers), all the way to apps whose purpose is to sell your data to whoever.

Read More

There’s plenty of advice about how to make your CV stand out when it’s reviewed by hiring managers, but what about the AI that pre-screens job applications these days?

Can anything be done in face of such advanced technologies?:

The time of spending hundreds of man-hours filtering through thousands of CVs and online job-board profiles for new employees is coming to an end. For example, ideally, a company, specializing in AI recruiting services, claims on their blog and estimates their AI candidate sourcing algorithm can “reduce time to hire from 34 days to 9 days”. This is a 73.53% increase in candidate sourcing and on boarding efficiency utilizing a non-biased process that removes stereotypes from sourcing and finds candidates that are technically appropriate for the position.

Assuming these AI systems work the same way as other AI systems in the wild, they are equivalent to subtracting points for “bad” keywords, adding points for “good” keywords and then summing up the result. Of course to qualify for “AI” it must be all hidden in a black box.

Read More

Such concentration of bad AI news is quite atypical these days, but here you go:

As time goes, more and more cracks are showing on the self driving car narrative. In June, one of the prominent startups in the competition - Drive.ai got acqui-hired by Apple, reportedly days before it would have ran out of cash. For those not well versed in startup valuation, this is not the best imaginable outcome.

Cruise is apparently plagued with glitches, one of Alphabet’s execs admitted that there has been much hype in the space, while Waymo’s valuation got pretty seriously (40% !!!) slashed by Morgan Stanley.

Read More

Oh wait, this time it’s not just AI, it’s the researchers, personally! Canadian Immigration, Refugee and Citizenship department can see right through this and will not let AI researchers steal Canadian jobs:

Mwiza Simbeye, who is from Zambia and is currently a Google research intern based in New York, was denied a visa to attend NeurIPS in Montreal last year. He says that in its denial letter, the Canadian embassy told him they were “not satisfied” that he would leave Canada once the conference was over.

Read More

How smart is the best AI really? The answer overwhelmingly depends on who you ask. Mostly because some focus on the strengths and others on the weaknesses. In the first school of thought you would get the headline “Pocket Calculator Shown To Have Superhuman Skills”, because it’s true that it surpasses human intelligence on a variety of tasks. In the AI world this is equivalent to the people amazed by AI winning over world chess and go champions. In this sense, AI has been more intelligent than people for quite some time.

But then, as noted in Rebooting AI, if you ask Google, the AI powerhouse, something like “Was Barrack Obama alive in 1994?”, it will give you a link to wikipedia telling you to go figure it out yourself, because it’s too hard for AI to come to this conclusion from the wikipedia by itself (Disclaimer: I work at Google Research, but not on anything mentioned in this post). Yet I’d guess that any chess champion would have no trouble figuring this out after reading the article. So who is smarter after all?

Read More

A part of what makes experience in medical profession so valuable, is that the experienced doctors can often effectively see a lot of clues just by looking at us and observing our behavior, not just listening to what we have to say to them. One of the clues for sure is our language. People in different psychosomatic states will use different vocabulary, style, and sentence structures. I, for example, tend to use shorter sentences when in strong pain, on average.

Read More

Bias is the topic of the day in AI, but bias have been here with us since forever. When modern statistics was born in breweries, bias in data wasn’t a big issue, but as soon as it entered the arena of what people are like and how they think, it had become the scary and elusive monster of the field. A textbook example of biased data ruining a survey is Ann Landers 1975 survey, which found that 70 percent of Americans would not have children, if they had to do it over again. The bias in that survey, as is the case in contemporary online surveys, stems from the fact that people who bother to respond are only those who feel strongly enough about the topic. Then you have a bias in your research: you thought you were learning something about the society at large, but instead only learned something about a bunch of weirdos who care respond to these surveys.

Read More

What is a language model? We used to play a game with a bunch of players around a table. You start a story on a piece of paper, just the first word. Then you pass it to the next person while you receive a story from the other neighbor. You write the next word and fold the paper so that only the last word is visible. Then you continue, always trying to write a good next word based on the last one. After a whole round of passing the stories, the papers are unrolled and the stories are read. Hilarity ensues. Sometimes.

Read More

There’s two types of crime: a classic crime that happens regardless of whether the police was there to see it, say a burglary or fraud. And then there is quantum crime that wouldn’t happen were there no police to see it. I mean the crime itself would happen, but it would never get into police data, so it’s almost as if it never happened in the first place. An example of this is speeding, biking without a helmet, this sort of thing.

With Predictive Policing there is an AI system telling police officers where to go. The system’s objective is to send the officers to places where the crime is most likely to occur. You can see how this can easily turn out really bad: PredPol sends the officers to a particular neighborhood, the officers respond to all sorts of crimes, the data is collected, and now when the system is retrained, it doubles down on sending the officers to the same place, because that’s where the crime happens.

For some reason this mechanism is called racial bias reinforcement and people are worried.

Read More

When you drive a car, there is quite a few things that can get in your way. For some of them it’s safer to just ignore them. These are birds, flying plastic bags or newspapers. There are other things which you learn to respect. Bisons, mooses, that sort of things.

Read More