Amazon’s facial ID incorrectly identifies members of Congress as criminals

Congress members want a meeting with Jeff Bezos to discuss Rekognition mistakes

amazon rekognition confuses congress criminals

Amazon may want to go back to the drawing board for this one. A facial recognition tool the retail giant currently offers developers made quite the mistake in a test conducted by the American Civil Liberties Union (ACLU), The New York Times reports. Rather than identifying 28 members of Congress as members of the country’s legislative branch, the tool instead classified these individuals as police suspects. To make matters worse, a disproportionate number of African-American and Latino Congress members were incorrectly identified, raising questions of how far racial profiling has embedded itself even in our machine learning software.

Now, Congress wants answers. Five of the misidentified lawmakers are requesting an “immediate” meeting with Amazon CEO Jeff Bezos. On Thursday, July 26, Reps. Jimmy Gomez, John Lewis, Luis Gutierrez, and Mark DeSaulnier, as well as Sen. Edward Markey, sent letters to the online retail giant asking why they were mismatched. But beyond the personal effects of these cases of mistaken identity, the congressmen are asking for answers as to “how to address the defects of this technology in order to prevent inaccurate outcomes.”

Amazon has not indicated yet whether its chief executive will accept these meetings.

In a separate statement, Rep. G.K. Butterfield noted, “I am troubled by the inaccurate outcomes associated with this technology, as there are clear blind spots that will have unintended consequences specifically for people of color. While this technology could have far-reaching economic potential, I encourage Amazon to better train its users on best practices for using this technology, be open and up-front about its limitations, and hire more employees of color who can properly assist with addressing the defects of this technology.”

Among the misidentified lawmakers were Reps. John Lewis and Bobby Rush, both of whom are well-known civil rights leaders and members of the Congressional Black Caucus. But instead of being recognized as elected representatives, these two men (along with 26 others), were falsely flagged as criminals. These mistakes occurred when the ACLU took Amazon’s software to analyze images of all members of Congress against a database of 25,000 publicly accessible mugshots. This resulted in the 28 Congress members being incorrectly identified as lawbreakers, which represents a 5 percent error rate.

amazon rekognition confuses congress criminals racial bias

“This test confirms that facial recognition is flawed, biased, and dangerous,” Jacob Snow, a technology and civil liberties lawyer with the ACLU of Northern California, told The New York Times.

In response to the test, Nina Lindsey, an Amazon Web Services spokeswoman, said that the ACLU used the face matching tool (called Amazon Rekognition) differently from Amazon’s recommendations. “It is worth noting that in real-world scenarios, Amazon Rekognition is almost exclusively used to help narrow the field and allow humans to expeditiously review and consider options using their judgment,” Lindsey said in a statement.

She further pointed out that the ACLU kept the tool’s “confidence threshold” at the default of 80 percent. This means that the group took into account any face matched as 80 percent similar to the police database. However, it’s worth pointing out that Amazon uses the same threshold in an example on its website, showing how employees’ faces can be matched with their ID badges. Lindsey said that in the case of police departments, Amazon recommends that a threshold of 95 percent similarity be implemented to avoid these sorts of mistakes.

amazon rekognition confuses congress criminals scan
Rep. Sanford Bishop (D-Ga.) was falsely identified by Amazon Rekognition as someone who had been arrested for a crime. ACLU

Amazon has previously come under fire for selling Rekognition. In May, 24 civil liberties groups, led by the ACLU, penned a letter to Amazon’s CEO Jeff Bezos requesting that the software cease being sold to law enforcement agencies. The groups expressed concerns that the software could be used to surveil protestors, immigrants, or general members of the public, rather than just police suspects. Amazon employees, investors, and academics have since made similar demands of Amazon.

For the time being, however, that doesn’t seem to be in the cards for the web giant. As Amazon’s spokesperson said, “We remain excited about how image and video analysis can be a driver for good in the world.”

Updated on July 27: Congress is demanding answers from Amazon regarding the mistakes made by its Rekognition software. 

Emerging Tech

Robots are going to steal 75 million jobs by 2025 — but there’s no need to panic

According to the World Economic Forum, robots and A.I. will take 75 million jobs from hardworking humans by 2025. That's the bad news. The good news is that they will create far more jobs than that.
Home Theater

What’s new on Amazon Prime Video (October 2018)

Amazon Prime Video adds new titles each month that are available for free to all Prime members. Check out our list to find all the content hitting Amazon Prime Video in September and October 2018, from new comedies to terrifying horror…

Cyber Monday 2018: When it takes place and where to find the best deals

Cyber Monday is still a ways off, but it's never too early to start planning ahead. With so many different deals to choose from during one of the biggest shopping holidays of the year, going in with a little know-how makes all the…
Emerging Tech

An A.I. is designing retro video games — and they’re surprisingly good

Researchers from Georgia Tech have demonstrated how artificial intelligence can be used to create brand-new video games after being shown hours of classic 8-bit gaming action for inspiration.
Smart Home

Amazon might open 3,000 cashier-free Amazon Go stores by 2021

According to new reporting by Bloomburg, anonymous sources within Amazon say that CEO Jeff Bezos is considering opening up to 3,000 of the company's cashier-less, experimental Amazon Go stores by 2021.
Emerging Tech

Wormlike motion sculptures show how athletes move in 3D

Researchers at MIT have developed a system that offers athletes a unique way to visualize their bodies in motion. An algorithm scans 2D videos of a person in motion, and generates data points that can be 3D-printed into "motion sculptures."
Emerging Tech

Harvard’s soft robotic exosuit adapts itself to the needs of every wearer

Harvard engineers have developed a new multi-joint, textile-based soft robotic exosuit, designed to help soldiers, firefighters, and other rescue workers. Here's what makes it so exciting.
Emerging Tech

These flying cars want to take your commute to new heights

The future is closer than you'd think: Companies around the world are working on flying car models, with many successful tests! Here are all the flying cars and taxis currently in development, and how they work!

Tap Strap wearable keyboard gains support for VR applications

TAP System's wearable keyboard gains support for virtual reality, now compatible with Windows Mixed Reality, Oculus Rift, and HTV headsets. Type and tap for up to eight hours in VR without needing to look at a physical keyboard.
Emerging Tech

Robot jellyfish could be used to patrol fragile coral reefs

Could schools of robotic jellyfish soon be patrolling the world’s oceans, monitoring fragile environments such as coral reefs? A team of United States researchers certainly thinks so.
Emerging Tech

Versatile robotic skin gives stuffed horse, other inanimate objects some giddyup

Researchers at Yale University have developed a new sensor-packed robot skin that can be wrapped around inanimate objects, such as toys, to transform them into functioning robots.
Emerging Tech

JackRabbot 2 is Stanford’s friendly new campus-roaming social robot

JackRabbot 2 is a robot developed by researchers at Stanford University -- designed to navigate around the campus, while carrying out friendly interactions with the humans around it.
Emerging Tech

New sustainable plan to mitigate climate change involves… a hot dog cooker?

Chemists have demonstrated a new, energy-efficient method of pulling carbon dioxide directly from the air. The secret ingredients? An air humidifier and a solar-powered hot dog cooker.
Emerging Tech

Removing ‘zombie cells’ in the brain could help battle the effects of dementia

Researchers at the Mayo Clinic have demonstrated how the removal of so-called "zombie cells" can help reverse the effects of dementia-style cognitive decline in mice. Here's what they did.