AI News, 'I was shocked it was so easy': meet the professor who says facial ... artificial intelligence
The invention of AI ‘gaydar’ could be the start of something much worse
(And to be clear, based on this work alone, AI can’t tell whether someone is gay or straight from a photo.) But the research captures common fears about artificial intelligence: that it will open up new avenues for surveillance and control, and could be particularly harmful for marginalized people.
One of the paper’s authors, Dr Michal Kosinski, says his intent is to sound the alarm about the dangers of AI, and warns that facial recognition will soon be able to identify not only someone’s sexual orientation, but their political views, criminality, and even their IQ.
This pseudoscience, physiognomy, was fuel for the scientific racism of the 19th and 20th centuries, and gave moral cover to some of humanity’s worst impulses: to demonize, condemn, and exterminate fellow humans.
The paper states: “Given a single facial image, [the software] could correctly distinguish between gay and heterosexual men in 81 percent of cases, and in 71 percent of cases for women.” These rates increase when the system is given five pictures of an individual: up to 91 percent for men, and 83 percent for women.
It guessed right 81 percent of the time for men and 71 percent of the time for women, but the structure of the test means it started with a baseline of 50 percent — that’s what it’d get guessing at random.
As Philip Cohen, a sociologist at the University of Maryland who wrote a blog post critiquing the paper, told The Verge: “People are scared of a situation where you have a private life and your sexual orientation isn’t known, and you go to an airport or a sporting event and a computer scans the crowd and identifies whether you’re gay or straight.
To try and ensure their system was looking at facial structure only, Kosinski and Wang used software called VGG-Face, which encodes faces as strings of numbers and has been used for tasks like spotting celebrity lookalikes in paintings.
They do test their AI on other picture sources, to see if it’s identifying some factor common to all gay and straight, but these tests were limited and also drew from a biased dataset — Facebook profile pictures from men who liked pages such as “I love being Gay,” and “Gay and Fabulous.” Do men in these groups serve as reasonable proxies for all gay men?
(“Study 1a showed that facial features extracted by a [neural network] can be used to accurately identify the sexual orientation of both men and women.') Experts say this is a misleading claim that isn’t supported by the latest science.
There may be a common cause for face shape and sexual orientation — the most probable cause is the balance of hormones in the womb — but that doesn’t mean face shape reliably predicts sexual orientation, says Qazi Rahman, an academic at King’s College London who studies the biology of sexual orientation.
As Matton writes in his blog post, approaches have ranged from “19th century measurements of lesbians’ clitorises and homosexual men’s hips, to late 20th century claims to have discovered ‘gay genes,’ ‘gay brains,’ ‘gay ring fingers,’ ‘lesbian ears,’ and ‘gay scalp hair.’” The impact of this work is mixed, but at its worst it’s a tool of oppression: it gives people who want to dehumanize and persecute sexual minorities a “scientific” pretext.
But on the other hand, it reinforces the devalued position of that kind of desire,” setting up hetrosexuality as the norm and framing homosexuality as “less valuable … a sort of illness.” And it’s when we consider Kosinski and Wang’s research in this context that AI-powered facial recognition takes on an even darker aspect — namely, say some critics, as part of a trend to the return of physiognomy, powered by AI.
In a widely shared rebuttal titled “Physiognomy’s New Clothes,” Google researcher Blaise Agüera y Arcas and two co-authors wrote that we should expect “more research in the coming years that has similar … false claims to scientific objectivity in order to ‘launder’ human prejudice and discrimination.” (Google declined to make Agüera y Arcas available to comment on this report.) Kosinski and Wang’s paper clearly acknowledges the dangers of physiognomy, noting that the practice “is now universally, and rightly, rejected as a mix of superstition and racism disguised as science.” But, they continue, just because a subject is “taboo,” doesn’t mean it has no basis in truth.
Kosinski says his research isn’t physiognomy because it’s using rigorous scientific methods, and his paper cites a number of studies showing that we can deduce (with varying accuracy) traits about people by looking at them.
So it will re-create, amplify, and continue on the trajectories we’ve taught it, which are always going to reflect existing cultural norms.” We’ve already created sexist and racist algorithms, and these sorts of cultural biases and physiognomy are really just two sides of the same coin: both rely on bad evidence to judge others.
Writing at the deep learning education site Fast.ai, researcher Jeremy Howard concludes: “It is probably reasonably [sic] to assume that many organizations have already completed similar projects, but without publishing them in the academic literature.” We’ve already mentioned startups working on this tech, and it’s not hard to find government regimes that would use it.
50 grand challenges for the 21st Century
ARTIFICIAL INTELLIGENCE danah boyd, Principal Researcher at Microsoft ResearchData-driven technologies are increasingly being integrated into many different parts of society, from judicial decision-making processes to automated vehicles to the dissemination of news.
Developing responsible sociotechnical systems will require bridging the social-technical gap that can easily emerge as social actors and technical actors speak past one another.
Missy Cummings, Professor, Humans and Autonomy Lab, Duke UniversityI think one of the most important challenges faced by robotic systems of the future, which include driverless cars, drones, surgical and manufacturing robots, is how will we be able to certify these systems as safe, particularly those that embed artificial intelligence?
By their very nature, artificial intelligence algorithms reason probabilistically and as uncertainty increases in the world, uncertainty increases in an algorithm’s ability to successfully and safely come to a solution.
Viktor Mayer Schonberger, Professor of Internet Governance and Regulation, Oxford Internet InstituteMy #1 issue is not the future of democracy (or related issues such as fake news, Trump, social networking bubbles, or even cybersecurity), but the future of humanity.
It's important that machine learning be researched openly, and spread via open publications and open source code, so we can all share in the rewards.
An intelligent agent (animal or robot) that can manipulate the physical world while sensing the results of said manipulation forms one half of a complex dynamical system.
Artificial intelligence is making some real progress right now, and our work is less to worry about a science fiction robot takeover, and more to see how technology can be used to help with human reflection and decisionmaking rather than to entirely substitute for it.
CITIES AND GLOBAL DEVELOPMENT Mary Barra, CEO, General MotorsThe auto industry stands at an inflection point where rapidly advancing technology and evolving customer needs offer a unique opportunity to transform our relationship with customers, communities and the environment.
Thanks to connectivity, electrification, autonomous vehicles and car- and ridesharing, the way customers interact with our vehicles is going to change in a way that hasn't happened since the industry was born more than 100 years ago.
of seeing more vernacular architecture overlapped with contemporary design is a desire to see cultural identities expressed as much as it is a desire to see climate adaptive solutions for space.
Vishaan Chakrabarti, Associate Professor of Practice at Columbia Graduate School of Architecture, Planning and PreservationThe major new challenge for the fields of architecture and urbanism will be to build what I call the 'Public Metropolis,' which means cities that are more ecologically sound, more equitable, more humane in their deployment of technology, more intense in their creation of new infrastructure, and more fervent in their roles as beacons for a free, diverse and open global society in a time when nativism and fascism are on the rise.
The debate of whether to build dense, transit-based cities as the most environmentally sound growth model in a world in which billions are reaching the middle class is largely settled: the question that remains is not whether to build better cities, but how.
Great civic architecture for both public and private projects will be pivotal to this question by enabling the creation of new cultural buildings, commercial projects, and infrastructures that read and write with the specifics of a place, so that we maintain local identities in a global world.
Rochelle Kopp, founder and Managing Principal of Japan Intercultural CounselingI would say that one of the biggest challenges for the 21st Century as relates to Japan and Asia, and indeed the rest of the world, is related to questions of immigration (which includes refugee issues).
These have of course received a lot of attention in the media, but the discussions are often stuck at a basic level, and governmental policies and programs are often not sufficiently addressing the issues.
There is a lot of room for further application of the lessons of the cross-cultural field in areas outside of business (where they are most often being utilised today), to help countries address issues related to immigrants and refugees.
Metropolitan Policy Program, Brookings InstituteThe real political and societal changes I’m seeing are taking place at the micro-local level: the biggest sociopolitical movement has been the organisation at the “place”
we are fundamentally inserting a new level of governance in society, and it’s taking the form of neighbourhood associations at the super local level, taking the form of improvement districts, special assessment districts, like in Midtown Manhattan.
Business improvement districts in particular are making leaps and bounds in the management of our society and they are recognising and working with technology firms to far better understand how these places work.
The issue is coming up with software that will create the mega database that will understand every part of the built environment at the place level, and eventually, the metropolitan level.
So when a city or when a business improvement district makes a major capital investment in the future, you could foresee the time that we’ll be able to say, ‘okay, let’s build the Second Avenue subway.
We’re getting closer to saying this is going to be the economic and fiscal benefit of doing that, and here are the unintended consequences we need to be concerned about: congestion, gentrification, displacement, whatever.
For African urbanisation to become a positive economic and social development, as opposed to a ticking time-bomb, urban planning needs to incorporate total populations, not simply the rich and middle classes;
Community participation in slum redevelopment initiatives has proven to be a far more productive and cheaper way of going about things than imposing ill-conceived, expensive schemes from above.
by that I mean not just the 1.3m that die on the roads each year (clearly unacceptable) but also the broader implications (effects on mental health and respiratory illness through poor air quality;
automated vehicles are one application of AI but what are the wider implications for employment (need for universal basic income?), privacy and security Shin-pei Tsay, ‎Executive Director, Gehl InstituteWithin urban areas, a significant constraint today and into the future will be how people move around the city.
Whatever technology may accomplish, we will still need to think about how space is used: automated and ride-sharing vehicles take up as much room as regular cars, whether they're on the road or parked off the street.
Going into the future, urban space still needs to be designed to maximize places for people to congregate, which are key to building social connections, fostering a sense of belonging, and encouraging community efficacy.
We even reduce economic limitations in labor markets when we plan for places that allow people to shorten their commute distances and have access to stores, schools, and other daily services.
Luke Alphey, visiting professor, Department of Zoology, University of OxfordAgricultural pest insects, and mosquitoes transmitting diseases, are long-standing problems for which we still have no satisfactory solution, indeed the problems are becoming more pressing.
One gene drive system, involving inserting into mosquito cells a large amount of foreign (to the mosquito) DNA in the form of an intracellular bacterium (Wolbachia), has entered field trials in several countries.
label and regulatory system by adroit marketing and some technicalities and perhaps illustrates what could be done if the field were not caught up in the baggage and polarised politics of the GM crops “debate”.
Recent research has shown that a country’s ratio of health to social service spending is predictive of some key health outcomes, like life expectancy, infant mortality, and maternal mortality.
More research is needed, to measure the health care cost savings of early childhood education or income support programs, and to identify the most sustainable integrated models.
genes so efficiently is being harnessed as a strategy to create new food, therapeutics, materials and methods for controlling the spread of diease. A
challenge moving forward is how to best engage the public with this fundamental science that really can positively impact human life and the world we live in.
I believe that we must continue to discuss and consider the profound societal and ethical impacts of CRISPR technology and ensure that it is not abused.c Joel Garreau, author, journalist, Professor of Law, Culture and Values, Sandra Day O’Connor College of Law, Arizona State UniversityThe major challenge consuming me is that the wheels are coming off the Enlightenment right now, on our watch, and it’s our own damn fault.
Greatest frustration: It is deeply annoying and vexing that CRISPR-cas9 and other gene editing techniques are being applied to treatment of rare diseases and a host of pharmacology development, but little investment is directed toward application of state-of-the-art gene editing or metagenomic sequencing and detection for point-of-care diagnostics creation.
There are even innovations that allow identification on-the-spot of infections with previously unknown microbes, based on conserved genetic regions found in classes of viruses or bacteria.
If Ebola broke out somewhere tomorrow we are better off today in that some methods for quickly identifying the virus in blood samples exist, but even now they remain noncommercial, require a laboratory and have no relevance to real-world conditions.
The goal on researchers' parts was to understand what genetic switches had to occur to turn a bird flu into a potentially catastrophic human airborne transmissible pandemic strain.
especially if it got into the wrong hands.That was then, this is now: The technology of gene modification is far more advanced, and application of cutting edge gene excision and incision techniques makes gain-of-function work potentially far easier, and more dangerous.
it could mean we can no longer safely carry out not only complex, lifesaving treatments such as chemotherapy and organ transplants but also more routine operations like caesareans and hip replacements.
More needs to be done to improve our ability to diagnose, treat and prevent drug resistant infections and to speed up development of new antibiotics to replace those no longer effective in protecting us against deadly infections.
Anit Mukherjee, policy fellow at the Center for Global DevelopmentTechnological innovation is progressing rapidly not only in the digital sphere but also in areas such as health, education, nutrition, food safety and life-saving/enhancing drugs.
In a world where there is a sentiment against movement of goods and people, how can developing societies adapt to increasing inequalities and build systems of governance to ensure human security?
Pardis Sabeti, Associate Professor of Organismic and Evolutionary Biology and of Immunology and Infectious Diseases, Harvard UniversityThe recent Ebola and Zika epidemics exposed our global vulnerabilities to deadly microbial threats and highlighted the need for proactive measures in advance of outbreaks and swift action during them.
Robert Sparrow, adjunct professor, Centre for Human Bioethics, Monash UniversityWhat does justice require of wealthy Northern states when confronted by mass migration from increasingly impoverished Southern countries as a result of accelerating climate change?
As technological developments increasingly drive social change, how can democratic societies empower ordinary people to have a say in the decisions that shape the technological trajectories that will in turn determine what the future looks like?
How can the public have meaningful input into the character of the algorithms that will increasingly determine both the nature of their relationships with other people on social media and their access to various important social goods?
Vast amounts of international travel, increasing urbanisation and a changing climates means that viruses can cross borders and spread around the globe faster than ever before.
Gavin Yamey, professor of the practice of global health, Duke University Global Health InstituteI believe one of the most urgent global issues that we face in 2017 and beyond, and one that we are woefully ill-prepared for, is the threat of epidemics and pandemics.
Cities are places where infrastructure gets locked in for decades, if not centuries, but city planners must make investments now in a world where technology is changing rapidly where people live, work and play, and how they access buildings, transport, energy and waste management.
Carey King, assistant director, University of Texas at Austin Energy InstituteWe need a discussion as to what political leaders, business leaders, and citizens think is an appropriate distribution of wealth across the entire population.
This focuses on the real question (how many people have what, independent of the size of the economy, though the two are linked) instead of discussing how to shape policies and taxes to achieve an unspecified growth target independent of wealth distribution.
Trump, Brexit, and Le Pen are representations that people understand growth only for the elite in the West is no longer tenable.An issue that has not received enough attention in the media and popular understanding is that the Earth is finite and this fact will have real world physical, economic, social, and political implications.
Neoclassical economics ignores this obvious fact, yet it is used to guide most policy (eg, economic projections and scenarios), including that for climate change mitigation.
We are already grappling with this problem across our developing member countries and with deteriorating river or surface water quality, lack of sufficient ground water sources and increasing dependence on sea water as a supply source, we have to bring in innovations in water management.
ADB is working with a large number of utilities to address these issues and as we engage on a long term basis with many cities and utilities, we will be actively exploring opportunities to bring in value for money propositions so that the utility benefits in the long term.
William Ryerson, founder and president, the Population Institute and Population Media CenterPerhaps a summary is that the human enterprise has outgrown the long-ability of the planet’s renewable resources to support us at our current numbers and our current rates of consumption and waste generation.
Technological improvements, while potentially important in reducing per capita impact, are not sufficient to make us sustainable unless we also stop growth in human numbers and reduce average consumption, while simultaneously lessening the gap between the richest and the poorest people on the planet.
For that reason, Population Media Center utilises long-running serialised dramas in various countries to create characters that gradually evolve into positive role models for the audience to bring about changes in social norms on a broad array of critical issues.
It used to be that in the summer it was a really quiet time for the grid operator compared to the winter, but now they are having this peak in generation in summer due to solar energy when demand is low.
When non-authoritative information ranks too high in our search results, we develop scalable, automated approaches to fix the problems, rather than manually removing these one-by-one.
Eddie Copeland, director of government Innovation at Nesta, a UK charity that has looked at the future of democracy in the digital worldRather than waiting for politicians to make decisions and then we all argue over whether what they say reflects reality, we could have tools that engage people much earlier in the process so they can be involved in formulating ideas and drafting legislation, following the course of how ideas go from concept to becoming laws and how effective they are in reality.
Nonny de la Pena, virtual reality journalist and CEO of Emblematic GroupCall me idealistic, but I really believe if you have an informed global citizenry, then people are going to make better decisions.
Stephan Lewandowsky, a psychologist at University of Bristol who studies persistence of misinformation in societyHaving a large number of people in a society who are mis-informed and have their own set of facts is absolutely devastating and extremely difficult to cope with.
Will Moy, director of Full Fact, an independent fact checking organisation based in the UKEven if we have structures that impose constraints on people in power and we put pressure on powerful people to be honest with us, in a sense, all of that is being circumvented by social media.
it is also the most powerful advertising platform there has ever been.We need a more equipped environment - we need watchdogs that will go around and say hang on, this doesn’t stack up and ask them to correct the record.
Paul Resnick, professor of information at the University of Michigan who developed a tool for identifying rumours on social media called RumourLensThe fundamental challenge we now face is how to handle a setting where anybody can get their views disseminated without intermediaries to prevent the distribution.
Victoria Rubin, director of the language and information technology research lab at Western University, Ontario, CanadaIf there are people who are willing to blatantly refuse to believe that something is a lie, no matter how hard you try, they won't listen.
The second newer misleading type of fakes that's gaining traction is native ads (specifically, in news), or sponsored content that's disguised as editorials, or what's formerly known as advertorials.
Why Deep Learning Is Suddenly Changing Your Life
Over the past four years, readers have doubtlessly noticed quantum leaps in the quality of a wide range of everyday technologies.
To gather up dog pictures, the app must identify anything from a Chihuahua to a German shepherd and not be tripped up if the pup is upside down or partially obscured, at the right of the frame or the left, in fog or snow, sun or shade.
Medical startups claim they’ll soon be able to use computers to read X-rays, MRIs, and CT scans more rapidly and accurately than radiologists, to diagnose cancer earlier and less invasively, and to accelerate the search for life-saving pharmaceuticals.
They’ve all been made possible by a family of artificial intelligence (AI) techniques popularly known as deep learning, though most scientists still prefer to call them by their original academic designation: deep neural networks.
Programmers have, rather, fed the computer a learning algorithm, exposed it to terabytes of data—hundreds of thousands of images or years’ worth of speech samples—to train it, and have then allowed the computer to figure out for itself how to recognize the desired objects, words, or sentences.
“You essentially have software writing software,” says Jen-Hsun Huang, CEO of graphics processing leader Nvidia nvda , which began placing a massive bet on deep learning about five years ago.
What’s changed is that today computer scientists have finally harnessed both the vast computational power and the enormous storehouses of data—images, video, audio, and text files strewn across the Internet—that, it turns out, are essential to making neural nets work well.
“We’re now living in an age,” Chen observes, “where it’s going to be mandatory for people building sophisticated software applications.” People will soon demand, he says, “ ‘Where’s your natural-language processing version?’ ‘How do I talk to your app?
The increased computational power that is making all this possible derives not only from Moore’s law but also from the realization in the late 2000s that graphics processing units (GPUs) made by Nvidia—the powerful chips that were first designed to give gamers rich, 3D visual experiences—were 20 to 50 times more efficient than traditional central processing units (CPUs) for deep-learning computations.
Its chief financial officer told investors that “the vast majority of the growth comes from deep learning by far.” The term “deep learning” came up 81 times during the 83-minute earnings call.
I think five years from now there will be a number of S&P 500 CEOs that will wish they’d started thinking earlier about their AI strategy.” Even the Internet metaphor doesn’t do justice to what AI with deep learning will mean, in Ng’s view.
Making China Great Again
His proposal for the 2018 budget would cut foreign assistance by forty-two per cent, or $11.5 billion, and it reduces American funding for development projects, such as those financed by the World Bank.
(The next day, in defiance of Trump’s threat, the resolution passed overwhelmingly.) To frame his vision of a smaller presence abroad, Trump often portrays America’s urgent task as one of survival.
In his first speech to the U.N., in September, Trump ignored its collective spirit and celebrated sovereignty above all, saying, “As President of the United States, I will always put America first, just like you, as the leaders of your countries, will always and should always put your countries first.” China’s approach is more ambitious.
In recent years, it has taken steps to accrue national power on a scale that no country has attempted since the Cold War, by increasing its investments in the types of assets that established American authority in the previous century: foreign aid, overseas security, foreign influence, and the most advanced new technologies, such as artificial intelligence.
If the initiative’s cost reaches a trillion dollars, as predicted, it will be more than seven times that of the Marshall Plan, which the U.S. launched in 1947, spending a hundred and thirty billion, in today’s dollars, on rebuilding postwar Europe.
Xi reiterated his support for the Paris climate deal and compared protectionism to “locking oneself in a dark room.” He said, “No one will emerge as a winner in a trade war.” This was an ironic performance—for decades, China has relied on protectionism—but Trump provided an irresistible opening.
“And the Chinese were going into every session and chortling about how they were now guarantors of the trading system.” By setting more of the world’s rules, China hopes to “break the Western moral advantage,” which identifies “good and bad” political systems, as Li Ziguo, at the China Institute of International Studies, has said.
Facing criticism for his lack of interest in global leadership, Trump, in December, issued a national-security strategy that singled out China and Russia and declared, “We will raise our competitive game to meet that challenge, to protect American interests, and to advance our values.” But, in his speech unveiling the strategy, he hailed his pullout from “job-killing deals such as the Trans-Pacific Partnership and the very expensive and unfair Paris climate accord.” The next day, Roger Cohen, of the Times, described the contradictions of Trump’s foreign policy as a “farce.” Some allies have taken to avoiding the Administration.
- On 23. januar 2021
'I was shocked it was so easy': meet the professor who says facial recognition can tell if you'r
'I was shocked it was so easy': meet the professor who says facial recognition can tell if you're gay Vladimir Putin was not in attendance, but his loyal ...
The A.I. Takeover Is Here! Should We Fear Artificial Intelligence?
John Lennox AI discussion starts at 22:30 - This is the shortened version. On Oct. 9, 2018, John Lennox addressed the critical questions surrounding artificial ...
2018 Isaac Asimov Memorial Debate: Artificial Intelligence
Isaac Asimov's famous Three Laws of Robotics might be seen as early safeguards for our reliance on artificial intelligence, but as Alexa guides our homes and ...
The Rise of AI
There's an AI revolution sweeping across the world. Yet few people know the real story about where this technology came from and why it suddenly took off.
Most AMAZING Robots In The World!
Check out the most amazing robots in the world! From artificial intelligence to futuristic walking humanoids, this top 10 list shows some of the coolest machines ...
The Public Policy Challenges of Artificial Intelligence
A Conversation with Dr. Jason Matheny Director, Intelligence Advanced Research Projects Activity (IARPA) Eric Rosenbach (Moderator) Co-Director, Belfer ...
The Future of AI with Kai-Fu Lee: Udacity Talks
Join us live for a rare conversation with Kai-Fu Lee, CEO of Sinovation Ventures, author of #AISuperpowers, and one of the leading pioneers in AI. He'll speak ...
Human Centered Artificial Intelligence with Justine Cassell, PhD
Justine Cassell, PhD gives a talk about how AI can learn from psychology and how psychological science or a human centered approach can improve outcomes ...
Microsoft AI – Everyday, Everywhere, for Everyone
Led by Harry Shum, EVP of AI and Research at Microsoft, presenters discuss everyday AI in Bing, Cortana, Office 365 and a partnership with Reddit announced ...
How to get empowered, not overpowered, by AI | Max Tegmark
Many artificial intelligence researchers expect AI to outsmart humans at all tasks and jobs within decades, enabling a future where we're restricted only by the ...