Your Artificial Intelligence is Not Bias-Free

Minimalist · Sep 12, 2017

Machines have no emotions. So, they must be objective—right? Not so fast. A new wave of algorithmic issues has recently hit the news, bringing the bias of AI into greater focus. The question now is not just whether we should allow AI to replace humans in industry, but how to prevent these tools from further perpetrating race and gender biases that are harmful to society if and when they do.
Click to expand...

https://www.forbes.com/sites/danielnewman/2017/09/12/your-artificial-intelligence-is-not-bias-free

RockLobster · Sep 12, 2017

The question now is not just whether we should allow AI to replace humans in industry...
Click to expand...

I wonder who the author thinks "we" are that we would have any say in the matter. People are a cost, offset against profits. That's all we are to the corporate machine that controls pretty much everything and "we" will be replaced whenever it is cost effective feasable to do so.
Bye bye checkout girl, hello self checkout. Bye bye manager, hello AI.

Minimalist · Sep 13, 2017

RockLobster said: ↑

Bye bye checkout girl, hello self checkout. Bye bye manager, hello AI.
Click to expand...

If I had to guess, I would say that checkout girl will be replaced much sooner than manager

RockLobster · Sep 13, 2017

Minimalist said: ↑

If I had to guess, I would say that checkout girl will be replaced much sooner than manager
Click to expand...

They already are being replaced. Regular computer systems replace manual workers all the time. If you go to Walmart in the evening, 1 real checkout girl, 15 automated checkouts. It doesn't seem to occur to these corporates when they replace people with technology that 15 less jobs is 15 less wages to be spent in the economy on their products and services.
Management is a little different, its tasks are not so repetitive so not so easily replaced by computer systems but they will be and that is the real future of AI in the workplace.
We spend centuries building a society that requires everyone to have a job for it to function, then greed steps in and says let's replace those jobs with technology.

guest · Jun 27, 2018

IBM hopes to fight bias in facial recognition with new diverse dataset
Biased facial recognition could affect everything from police work to hiring
June 27, 2018
https://www.theverge.com/2018/6/27/17509400/facial-recognition-bias-ibm-data-training

dBias is a big problem in facial recognition, with studies showing that commercial systems are more accurate if you’re white and male. Part of the reason for this is a lack of diversity in the training data, with people of color appearing less frequently than their peers.

The first dataset contains 1 million images and will help train systems that can spot specific attributes, like hair color, eye color, and facial hair.

The second dataset is the more interesting one. It’s smaller than the first, containing 36,000 pictures, but the faces within are an equal mix ethnicities, genders, and ages. In the same way that the facial attribute tags help train AI systems to recognize these differences, having a diverse mix of faces should help systems overcome various biases.

“This dataset [...] should really help designers to tune their algorithms,” said Puri. “Data is the foundation of AI, and it’s not just about building our own capabilities but the community’s as well.”
Click to expand...

itman · Jun 28, 2018

RockLobster said: ↑

If you go to Walmart in the evening, 1 real checkout girl, 15 automated checkouts. It doesn't seem to occur to these corporates when they replace people with technology that 15 less jobs is 15 less wages to be spent in the economy on their products and services.
Click to expand...

This happened at my local store shortly after Walmart announced they were raising the minimum starting wage. Assumed the self-checkout bit was previously corp. planned and approved to offset those wage increases. You will never beat the corps. when it comes to profits. ~ Removed Political OT Remarks ~

Anyway, I read the AI article in my morning's newspaper. All like stuff reminds me of a sci-fi book I read a few decades ago. It was about the last human on earth. Society delegated everything to machines. This caused society to get so lazy, they stopped maintaining the machines. Of course, everything stopped working which caused everyone to cease to exist; except for this poor SOB.

guest · Sep 20, 2018

IBM launches tool aimed at detecting AI bias
September 19, 2018
https://www.bbc.co.uk/news/technology-45561955

For example, in the past, image recognition systems have failed to identify non-white faces.

However, as they increasingly make automated decisions about a wide variety of issues such as policing, insurance and what information people see online, the implications of their recommendations become broader.

The IBM cloud-based software will be open-source, and will work with a variety of commonly used frameworks for building algorithms.

Customers will be able to see, via a visual dashboard, how their algorithms are making decisions and which factors are being used in making the final recommendations.
Click to expand...

Blog entry:
Introducing AI Fairness 360
September 19, 2018
https://www.ibm.com/blogs/research/2018/09/ai-fairness-360/

We are pleased to announce AI Fairness 360 (AIF360), a comprehensive open-source toolkit of metrics to check for unwanted bias in datasets and machine learning models, and state-of-the-art algorithms to mitigate such bias. We invite you to use it and contribute to it to help engender trust in AI and make the world more equitable for all. [...]
Click to expand...

guest · Sep 28, 2018

Majority of Orgs Failing to Make Machine Learning Fair, Safe & Balanced
September 27, 2018
https://www.infosecurity-magazine.com/news/orgs-failing-tmachine-learning/

The firm conducted its research among 2000 senior business leaders in the EU, discovering that over half (55%) of EU businesses have not included privacy provisions in their model-building checklist, whilst 53% do not account for compliance and 62% don’t include fairness and bias.

Only 14% of those polled accounted for compliance, privacy, fairness and bias in their model-building checklist, and O’Reilly Media warned that failing to do so will result in failed results from flawed, biased and unethical applications that could also put people’s privacy at risk.

Lorica argued that, too often, the task of developing machine learning technology falls to data scientists without insight from lawyers, compliance and privacy experts.
Click to expand...

guest · Sep 30, 2018

Unbiased algorithms can still be problematic
It depends on how they're used
September 30, 2018
https://techcrunch.com/2018/09/30/unbiased-algorithms-can-still-be-problematic/

Creating unbiased algorithms is a matter of having enough accurate data. It’s not about just having enough “pale males” in the model, but about having enough images of people from various racial backgrounds, genders, abilities, heights, weights and so forth.

But the further you get from pale males — adding women, people from different ethnicities, and so forth — “the harder it is for AI systems to get it right, or at least the confidence to get it right,” Brackeen said.

But in the case of law enforcement, no matter how accurate and unbiased these algorithms are, facial recognition software has no business in law enforcement, Brackeen said. That’s because of the potential for unlawful, excessive surveillance of citizens.

And even if the police were to have perfect information about every crime committed, in order to build a fair machine learning system, “we would need to live in a society of perfect surveillance so that there is absolute police knowledge about every single crime so that nothing is excluded,” he said.
Click to expand...

guest · Oct 10, 2018

Amazon scrapped 'sexist AI' tool
October 10, 2018
https://www.bbc.co.uk/news/technology-45809919

The artificial intelligence system was trained on data submitted by applicants over a 10-year period, much of which came from men, it claimed.

Reuters was told by members of the team working on it that the system effectively taught itself that male candidates were preferable.

"They literally wanted it to be an engine where I'm going to give you 100 resumes, it will spit out the top five, and we'll hire those," said one of the engineers who spoke to Reuters.

The system started to penalise CVs which included the word "women". The program was edited to make it neutral to the term but it became clear that the system could not be relied upon, Reuters was told.
It is not the first time doubts have been raised about how reliable algorithms trained on potentially biased data will be.
Click to expand...

guest · Nov 16, 2018

Most Americans believe algorithms will always be biased
November 16, 2018
https://www.engadget.com/2018/11/16/study-shows-americans-believe-algorithms-biased/

Pew has conducted a survey indicating that 58 percent of American adults believe algorithms and other programming will always contain some kind of human bias.

Why the skepticism? To some extent, it could stem from social networks. A clear 74 percent of study participants didn't think social media accurately reflected society. It pushed their emotional hot buttons (88 percent were at least sometimes amused, 71 percent angry) and frequently led to heated discussions whether or not they had all the facts.

The survey also suggested that most Americans didn't believe algorithms should be used in situations with far-reaching consequences. [...] Many didn't believe algorithms could accurately reflect complex human nature, and that the algorithms were both unfair and could violate privacy.
Click to expand...

guest · Dec 6, 2018

Microsoft Calls for Laws to Prevent Bias in Facial Recognition AI
December 6, 2018
https://www.bloomberg.com/news/arti...laws-to-prevent-bias-in-facial-recognition-ai

Microsoft Corp. called for new legislation to govern artificial intelligence software for recognizing faces, advocating for human review and oversight of the technology in critical cases.

Sellers of the technology must “recognize that they are not absolved of their obligation to comply with laws prohibiting discrimination against individual consumers or groups of consumers,” he added.
Smith also wants laws to require sellers of the products to explain what they do clearly and open up their services to testing by outside parties for accuracy and bias.

Microsoft thinks current facial recognition technology increases the risk of biased outcomes, can lead to new intrusions into privacy and can encroach on democratic freedoms if used for mass government surveillance, the executive added.
“We must ensure that the year 2024 doesn’t look like a page from the novel 1984,” Smith wrote.
Click to expand...

guest · Jan 20, 2019

Facebook backs an independent AI ethics research center
January 20, 2019
https://www.engadget.com/2019/01/20/facebook-tum-ai-ethics-research-center/

The social network has teamed up with the Technical University of Munich to back the formation of an independent AI ethics research center.

The plainly titled Institute for Ethics in Artificial Intelligence will wield the university' academic resources to explore issues that "industry alone cannot answer," including those in areas like fairness, privacy, safety and transparency.
Facebook is providing an initial grant of $7.5 million over the space of five years, and says it might also share its own findings and tools.
Click to expand...

guest · Jan 24, 2019

Microsoft Seeks to Restrict Abuse of its Facial Recognition AI
January 23, 2019
https://www.bloomberg.com/news/arti...o-restrict-abuse-of-its-facial-recognition-ai

“We do need to lead by example and we’re working to do that,” Microsoft President and Chief Legal Officer Brad Smith said in an interview, adding that some other companies are also putting similar principles into place.

“It would certainly restrict certain scenarios or uses,” Smith said of the principles, adding that Microsoft wouldn’t necessarily reject providing governments with the technology.
The company only wants to prevent law enforcement from using the technology for ongoing surveillance of a specific individual without the preferred safeguards, he said.

The company has turned down contracts for that reason, he said. [...] Another was deployment by a law enforcement agency in the U.S. that “we thought would create an undue risk of discrimination.”
Click to expand...

guest · Jan 29, 2019

mood said: ↑

IBM hopes to fight bias in facial recognition with new diverse dataset
Biased facial recognition could affect everything from police work to hiring
June 27, 2018
https://www.theverge.com/2018/6/27/17509400/facial-recognition-bias-ibm-data-training
Click to expand...

IBM releases Diversity in Faces, a dataset of over 1 million annotations to help reduce facial recognition bias
January 29, 2019
https://venturebeat.com/2019/01/29/...tions-to-help-reduce-facial-recognition-bias/

The company today published a research paper and accompanying dataset — Diversity in Faces (DiF), annotations of 1 million human facial images sourced from the publicly available YFCC-100M Creative Commons dataset — that aims to reduce prejudicial predictions in AI face-detecting algorithms.

“Face recognition is a long-standing challenge in the field of artificial intelligence (AI),” the authors of the paper wrote. “The goal is to create systems that detect, recognize, verify, and understand characteristics of human faces. [...]"
In preliminary tests, the scientists found that the DiFdata set provided a “more balanced” distribution and “broader coverage” of facial images compared to previous datasets.

“We believe by extracting and releasing these facial coding scheme annotations on a large dataset of 1 million images of faces, we will accelerate the study of diversity and coverage of data for AI facial recognition systems to ensure more fair and accurate AI systems,” Smith said.
Click to expand...

Research paper (PDF): https://www.research.ibm.com/artificial-intelligence/trusted-ai/diversity-in-faces/documents/Diversity-in-Faces-Publication.pdf

guest · Feb 13, 2019

The Real Reason Tech Struggles With Algorithmic Bias
February 12, 2019
https://www.wired.com/story/the-real-reason-tech-struggles-with-algorithmic-bias/

Are algorithms and artificial intelligence inherently prejudiced? Do Facebook, Google, and Twitter have political biases? Those answers are complicated.
But if the question is whether the tech industry doing enough to address these biases, the straightforward response is no.

This is exactly why the tech industry needs to actually invest in real cognitive bias training and empower true experts to address these issues, as opposed to spouting platitudes. Countering bias takes work

No matter how trained or skilled you may be, it is 100 percent human to rely on cognitive bias to make decisions.
Becoming overly reliant on data—which in itself is a product of availability bias—is a huge part of the problem.
Click to expand...

guest · Mar 7, 2019

AI Researchers Are Pushing Bias Out of Algorithms
March 6, 2019
https://www.bloomberg.com/opinion/a...ing-bias-out-of-algorithms?srnd=technology-vp

The perception that AI researchers and developers care more about algorithms and robots than about people is misguided.

To address issues of bias and unfairness, researchers are tackling two related challenges.
The first challenge is to spell out what “fairness” means in mathematical terms. This may seem difficult, but there’s actually an embarrassment of riches: Dozens of proposals are under consideration.
The second challenge is to put those abstract fairness definitions to practical use.

Still, there are danger zones.
Looking further into the future, the existence of a superintelligent system that doesn't have humanity’s best interests at heart — a well-loved science fiction theme — is becoming a credible possibility
Click to expand...

guest · Aug 25, 2019

The risks of amoral A.I.
The consequences of deploying automation without considering ethics could be disastrous
August 25, 2019
https://techcrunch.com/2019/08/25/the-risks-of-amoral-a-i/

Artificial intelligence is now being used to make decisions about lives, livelihoods, and interactions in the real world in ways that pose real risks to people.

It’s not that surprising with cars that (sometimes and under certain conditions) drive themselves and software that beats humans at games like chess and Go. You can’t blame people for being impressed.
AI is being used in a surprising number of applications, making judgments about job performance, hiring, loans, and criminal justice among many others. Most people are not aware of the potential risks in these judgments. They should be.

Physical risks pose an obvious danger, but there has been real harm from automated decision-making systems as well. AI does, in fact, have the potential to beneﬁt the world. Ideally, we mitigate for the downsides in order to get the beneﬁts with minimal harm.
In many other cases, individuals don’t even know that a choice not to hire, promote, or extend a loan to them was informed by a statistical algorithm.
Click to expand...

As a matter of fact, Reuters reports that other companies are blithely moving ahead with AI for hiring. A third-party vendor selling such technology actually has no incentive to test that it’s not biased unless customers demand it, and as I mentioned, decision makers are mostly not in a position to have that conversation. Again, human bias plays a part in hiring too. But companies can and should deal with that.
Click to expand...

guest · Jan 27, 2020

mood said: ↑

IBM releases Diversity in Faces, a dataset of over 1 million annotations to help reduce facial recognition bias
January 29, 2019
Research paper (PDF): https://www.research.ibm.com/artificial-intelligence/trusted-ai/diversity-in-faces/documents/Diversity-in-Faces-Publication.pdf
Click to expand...

Little grouse on the prairie: IBM's AI facial-recognition training dataset gets it in trouble... in Illinois
Photo subjects fling class-action sueball
January 27, 2020
https://www.theregister.co.uk/2020/01/27/ibms_facial_recognition_software_gets_it_in_trouble_again/

A class-action lawsuit filed late last week has accused IBM of using photos of millions of people in Illinois without informing them to build a facial-recognition dataset.

In the filing (Bloomberg Law has a copy here), lead plaintiff Tim Janecyk alleges that IBM violated the state's Biometric Information Privacy Act (BIPA) laws by using at least seven of his photographs from Flickr without informing him or the subjects of his photos.
The "Diversity in Faces" dataset comprises millions of images pulled from Flickr, a popular image-sharing site, under the Creative Commons licence.

Janecyk and the other members of the class have asked IBM to cough up $5,000 for every photo they used without consent.
Click to expand...

Log in or Sign up

Your Artificial Intelligence is Not Bias-Free

Minimalist Registered Member

RockLobster Registered Member

Minimalist Registered Member

RockLobster Registered Member

guest Guest

itman Registered Member

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

Log in or Sign up

Your Artificial Intelligence is Not Bias-Free

Minimalist Registered Member

RockLobster Registered Member

Minimalist Registered Member

RockLobster Registered Member

guest Guest

itman Registered Member

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

Useful Searches