After reading Make Your Job Application Robot-Proof, I thought I would write a post about a technology I have personally seen–AI interviews.
When I applied for a software engineering internship at JPMorgan Chase, I was required to complete a Pymetrics assessment and a HireVue interview. Recently, both of these technologies have received criticism for using proprietary algorithms that have not been audited by third parties to evaluate job candidates.
Pymetrics provides a service to recruiters that allows them to gain more insight into their candidates. A candidate plays brain games and Pymetrics generates a report that includes things such as personality traits and whether they are willing to take risks. However, the algorithm is not open source (the code is not public). While Pymetrics has released their own auditing framework, there is no proof that their algorithm has actually been audited with it, or by a third party. Dipayan Ghosh, a former privacy and public policy official at Facebook who now works with a think tank, thinks that Pymetrics’ internal audits are not sufficient. “We know of too many past cases where all a company needed to do is to self-certify, and it was shown to be perpetuating harms to society and, specifically, certain people. … The public will have little knowledge as to whether or not the firm really is making biased decisions if it’s only the firm itself that has access to its decision-making algorithms to test them for discriminatory outcomes.”
HireVue, a video interview company, has recently added AI job assessments. According to Meredith Whittaker, co-founder of the AI Now Institute at NYU, HireVue’s models are based on top employees at a firm. During an interview, the candidate’s emotions and face are monitored. Using this data, HireVue attempts to evaluate candidates based on the models of the top employees. Whittaker expresses similar concerns to those expressed by Ghosh about Pymetrics: “This is alarming, because firms that are using such software may not have diverse workforces to begin with, and often have decreasing diversity at the top. And given that systems like HireVue are proprietary and not open to review, how do we validate their claims to fairness and ensure that they aren’t simply tech-washing and amplifying longstanding patterns of discrimination?”
3D printed guns are becoming frighteningly common and popular. One needs to find a schematic for a gun, and a 3D gun can be printed without a license, background check or serial number to trace the gun if misused. US District Judge Robert Lasnik in Seattle issued a restraining order that interrupted plans of an activist to release gun designs for free on the internet. After the bill was passed, this activist found a way to release these designs another way. He started selling them to people who wanted to buy it. With Trump’s gun-friendly agendas, this is becoming inevitable. With the improving 3D technology, experts say that gun will be cheap, accessible and untraceable. The reason this is delaying is that, right now, 3D printing is expensive and quality needs to be really good to be able to produce a product like a gun. But over time, cost and quality will become cheaper and better.
“It’s bad enough that in the majority of states you can purchase a traditional firearm without undergoing a background check to ensure you are not prohibited. The availability of 3D-printed guns creates yet another loophole through which prohibited individuals could more easily obtain firearms” says Cassandra Crifasi, from the Johns Hopkins Center for Gun Policy and Research. Experts are seriously concerned with how much damage can be done with the 3D printed guns and people who misuse it can easily get away.
What is terrifying is that once 3D printing is cheaper and accessible to almost everyone, it will be next to impossible to stop people from making any product they wish to make. For example, a person who is banned to purchase firearms due to having past criminal records, can easily work their way around.
Digital Teleportation has began to cross many minds but definitely does not exist yet. This idea of digital teleportation combines machines (similar to drones), virtual reality, and fast wireless networks. Digital Teleportation is trying to change the way people experience the world, obviously just better. Although digital teleportation is yet to exist, some similar forms are starting to appear such as combinations of virtual reality, augmented reality alongside the use of drones.
Augmented Reality, Virtual Reality, and the use of drones are major concepts that were discussed periodically throughout class. In my opinion, the idea of digital teleportation is awesome but I do believe that it will come with multiple negative effects. I think having a 360 view of the location of wherever you want to be is an amazing technological advancement that definitely did not exist a couple years back. Drones, I believe are only beneficial. They can be used for so many things such as safety/protection as well as monitoring and scaling places before actually going there.
People looking for jobs spending so much time tweaking and enhancing their resumes only to submit themselves and never hear back from the company in which they submitted their resumes. Who would have thought that robots are the reason humans are never hearing back from certain companies who have job opportunities available. Employers save so much money and time using these robots. There is also a potential that using these robots can eliminate bias which is very beneficial. Using these robots with relation to college students is a very trendy topic that has views on both sides.
In my opinion, artificial intelligence is only going to enhance more and more as time progresses. The fact that employers trust these robots with actual people’s resumes is somewhat astonishing. However, this is so due to large technological advancements which have made such things like this possible. I think that robots can be used for different purposes rather than resume reviewing because people need to be working in order to support themselves meaning that a robot, who might have a technical deficiency, never gives a certain person a chance with the employer due to a resume issue.
* comment on axj5293 post on chip cards vs. magnetic stripes
Many think the chip credit card is a lot slower than the magnetic strip credit cards however, cards with a chip prove to be a lot more secure. As stated in the post, the chip credit cards are difficult to copy since there is new technology that encrypts the codes sent out from the credit cards. On the cards that utilize a chip, the chip refers to a computer microchip that automatically records every transaction and sends verification to the user. Card production companies decided to start designing credit cards with microchips for security reasons, making it less likely your card will be used to make fraudulent transactions. The United States government doesn’t require cards to have chips in it like in other countries. “U.S. merchants have foregone updating their payment technology to accept chip cards — leaving them more susceptible to credit card fraud.” I think that the United States needs to have a law that requires chip credit cards to ensure safety.
We talked about in class on how IoT is going to be used and applied to many new technologies from now to the future. However, IoT is going to cause issues as it will bring up many new disruptive innovations which will negatively impact other business like how the emergence of Uber negatively impacted the taxi industry. The Internet of things is not a disruptive innovation itself, but it will enhance technologies and are expected to be a part of the disruptive innovations. Some of the technologies that could become disruptive innovation are, robotics, A.I., Autonomous vehicles and more.
Recently Google has released incognito mode on google maps for iOS. This feature had already been released for Androids earlier this year. When you have this feature turned on you “can search for a location or navigate somewhere, and Google won’t save that information to your Google account.” In the article on The Verge it also stated that Google has announced that next month they will be releasing an update for Androids where they can mass delete information from their Google Maps timeline and location history. I think that what Google is doing hits on a major part of technology: privacy. Privacy is and always will be a huge issue, especially as more and more parts of our lives involve technology. Google is smart because they realize more people are worrying about what information is being tracked and who can access this information. This move made by Google is in the long run creating better customer relations with the millions of people who use their service every day.
An article by BBC recently talked about the interesting connection between humans and smart speakers. It claimed how if the speaker voice was too robotic, consumers would expect less of it, but if it sounded too human, then their expectations were very high as far as capabilities go. Oftentimes more advanced smart speakers, like one Google has recently worked on, are so realistic to human conversation and in turn have very advanced things that they can do. The article also talked about the decision of picking a voice for the speakers, and how the majority of the speakers use a female voice which can create more sexist behavior among listeners, especially when the speaker does not work properly. I found it to be very interesting how much thought and attention to detail is required for something as simple as the smart speaker’s voice, as it may seem irrelevant compared to the complexity of the technology and artificial intelligence. There is so much to consider when creating a product like this, and it is very impressive how companies think and research so much into these things.
While we talk a lot about design thinking when it comes to things like technology or projects on the more creative side, this Forbes article talks about how design thinking can be used in the laboratory setting as well. It shares the story of an employee, Lisa Ruiz, of a biopharmaceutical company, and how she has used the design thinking process to reach medical discoveries. She talks about meeting with different patients and using empathy to be able to find a treatment for their disease(s). She would listen to their stories and experiences and fears in order to better create a solution. I found this interesting because I always thought of design thinking to be something with a lot of imagination, and never would have thought that it could be beneficial in the scientific world. Ruiz says, in regards to treatment research, “It goes beyond a pill or injection; it goes to treating the whole person,” which I think really shows how design thinking can be a beneficial tool for people working on almost any project. By getting to understand what it is really like for people to live with a certain condition, Ruiz could work with a better mindset and have a different kind of motivation and understanding to reach more success.
During the summer, Facebook Inc., created an artificial intelligence program that had the ability to beat 5 poker players in a Texas Hold’em Tournament, at once! In the past, artificial intelligence programs have had the ability to beat a poker player one-on-one, however, this is a big step up considering the uncertainty of poker. AI has to consider strategies in poker like bluffing while battling against live opponents, especially because each opponent contains information about the cards that the other does not (what they have in their hand). Now, I can understand why this may seem like a silly venture for Facebook to undergo considering the issues they have with explicit content on their website, “fake news,” and the numerous privacy issues that have come to light over the past couple years. However, there are practical applications to this technology. The group that created it, which includes a research group from Carnegie Mellon University as well, is looking to apply this technology in other fields. Since poker is such an intricate game, the act of considering other users information and balancing this with a strategy to win is what the researchers strive to apply to various technologies. We have begun to see self-driving cars breach the market, and perfecting this type of artificial technology could be a step in making AI more prevalent in society. What other types of technical applications could you see a technology like this being used for?
“An Artificial Intelligence Program Just Beat 5 Poker Professionals in a Texas Hold’em Tournament:” https://time.com/5624968/facebook-ai-beats-professional-cardplayers/