Connect with us

Ai

I’m the Google whistleblower. The medical data of millions of Americans is at risk | Anonymous

Published

on

When I learned that Google was acquiring the intimate medical records of 50 million patients, I couldnt stay silent

I didnt decide to blow the whistle on Googles deal, known internally as the Nightingale Project, glibly. The decision came to me slowly, creeping on me through my day-to-day work as one of about 250 people in Google and Ascension working on the project.

When I first joined Nightingale I was excited to be at the forefront of medical innovation. Google has staked its claim to be a major player in the healthcare sector, using its phenomenal artificial intelligence (AI) and machine learning tools to predict patterns of illness in ways that might some day lead to new treatments and, who knows, even cures.

Here I was working with senior management teams on both sides, Google and Ascension, creating the future. That chimed with my overall conviction that technology really does have the potential to change healthcare for the better.

But over time I grew increasingly concerned about the security and privacy aspects of the deal. It became obvious that many around me in the Nightingale team also shared those anxieties.

After a while I reached a point that I suspect is familiar to most whistleblowers, where what I was witnessing was too important for me to remain silent. Two simple questions kept hounding me: did patients know about the transfer of their data to the tech giant? Should they be informed and given a chance to opt in or out?

The answer to the first question quickly became apparent: no. The answer to the second I became increasingly convinced about: yes. Put the two together, and how could I say nothing?

So much is at stake. Data security is important in any field, but when that data relates to the personal details of an individuals health, it is of the utmost importance as this is the last frontier of data privacy.

With a deal as sensitive as the transfer of the personal data of more than 50 million Americans to Google the oversight should be extensive. Every aspect needed to be pored over to ensure that it complied with federal rules controlling the confidential handling of protected health information under the 1996 HIPAA legislation.

Working with a team of 150 Google employees and 100 or so Ascension staff was eye-opening. But I kept being struck by how little context and information we were operating within.

What AI algorithms were at work in real time as the data was being transferred across from hospital groups to the search giant? What was Google planning to do with the data they were being given access to? No-one seemed to know.

Above all: why was the information being handed over in a form that had not been de-identified the term the industry uses for removing all personal details so that a patients medical record could not be directly linked back to them? And why had no patients and doctors been told what was happening?

I was worried too about the security aspect of placing vast amounts of medical data in the digital cloud. Think about the recent hacks on banks or the 2013 data breach suffered by the retail giant Target now imagine a similar event was inflicted on the healthcare data of millions.

I am proud that I brought this story to public attention. Since it broke on Monday several Congress members have expressed concerns including the Democratic presidential candidate Senator Amy Klobuchar of Minnesota who said the deal raised serious privacy concerns.

A federal inquiry has been launched into whether HIPAA protections have been fully followed.

I can see the advantages of unleashing Googles huge computing power on medical data. Applications will be faster; data more accessible to doctors; new channels will be opened that might in time find cures to certain conditions.

But the disadvantages prey on my mind. Employees at big tech companies having access to personal information; data potentially being handed on to third parties; adverts one day being targeted at patients according to their medical histories.

Id like to hope that the result of my raising the lid on this issue will be open debate leading to concrete change. Transfers of healthcare data to big tech companies need to be shared with the public and made fully transparent, with monitoring by an independent watchdog.

Patients must have the right to opt in or out. The uses of the data must be clearly defined for all to see, not just for now but for 10 or 20 years into the future.

Full HIPAA compliance must be enforced, and boundaries must be put in place to prevent third parties gaining access to the data without public consent.

In short, patients and the public have a right to know whats happening to their personal health information at every step along the way. To quote one of my role models, Luke Skywalker: May the force be with you.

Read more: https://www.theguardian.com/commentisfree/2019/nov/14/im-the-google-whistleblower-the-medical-data-of-millions-of-americans-is-at-risk

Ai

Of course Facebook and Google want to solve social problems. Theyre hungry for our data | Nathalie Olah

Published

on

By

Giving big tech companies power over the NHS or the climate crisis wont build a fairer world. But public ownership would, says author Nathalie Olah

We hear it said all the time, most recently in a national campaign for BT: Technology will save us. The slogan was plastered on billboards across the country as part of BTs new advertising campaign, linked to a UK-wide digital skills movement developed partly with Google. The sentiment is so ubiquitous that it even led to a dispute with a startup of a similar name. But in an era dominated by the big four (Google, Amazon, Facebook and Apple) the idea that tech will save us rings hollow, an example of utopian messaging being used to conceal the simple pursuit of profit.

Having proposed solutions to everything from food shortages to suicide prevention to climate breakdown, companies such as Google and Facebook two of the leading western companies in the artificial intelligence arms race claim theres almost nothing that cannot be tackled through tech. But there are reasons to be sceptical. These companies business models depend on the development of ever more complex algorithms, sustained by enormous quantities of data. This data is used to improve the algorithms but access to it is also sold to advertisers and third-party businesses.

Conquering new sources of data has therefore become their primary mission. And thats why theyre eyeing our public commons: telecommunications, energy and even urban space, which continuously generate enormous quantities of real-time data. In 2017, it was reported that Googles AI outfit DeepMind was in talks with the National Grid. DeepMinds founder, Demis Hassabis, expressed an interest in expanding technology similar to that used to minimise energy wastage at Google data centres where electricity usage had been cut by 15% across the energy grid. This is an improvement, of course, but as one of the main examples of how AI systems might be used to tackle climate change it is hardly inspiring.

It also neglects to mention that unprecedented access to our critical infrastructure and publicly generated data would be given to a US tech giant. The collaboration between DeepMind and the (privatised, shareholder-paying) National Grid has for now been abandoned for reasons that are unclear. A recent article in Forbes speculates that the two companies couldnt reach an agreement on costs and intellectual property rights, in perhaps the most telling example of big techs ambitions to boost revenues through the commandeering of national infrastructure. Could Googles recent engagement with BT be built on a similar ambition?

Giving tech giants the power to solve social problems would mean granting them an immense stake in almost everything that our society requires in order to function. Google is currently signing contracts with the NHS to process patient records, despite there being legal question marks over a similar arrangement with a London hospital a few years ago. Whats more, the climate crisis is a political, not a technological problem. Whatever improvements Google or Facebook could make to our infrastructure would still fall far short of solving it. And when environmental collapse stands to affect poorest communities the hardest, the question remains as to how an industry that drives extreme wealth inequality can really be said to help build a greener, more humane, world.

These companies are able to make it seem as though their sole ambition is to optimise and improve their systems for the greater good. But this rhetoric distracts us from the fact that they are ushering in a new kind of surveillance capitalism, whereby a small number of entities extract enormous amounts of wealth through their access to data that is generated by us, the public.

To ensure that we retain the control to manage these systems, and to avoid an unprecedented level of power and wealth being concentrated in the hands of a very small elite, our infrastructure urgently needs to be brought under state control. This is why the Green New Deal, backed by Alexandria Ocasio-Cortez in the US and taken up in the UK by the Labour party, is so important. Not only will the UK version pursue efforts to keep global average temperature rises below 1.5C, but by encompassing public ownership of energy companies it provides a democratic line of defence against the predations of Silicon Valley. Labours proposal to part-nationalise BT opens up a new front in this battle especially since the party is planning to help pay for it with a tax on big tech.

In the years to come, this will give the state a far stronger negotiating position on resources, both digital and physical, as well as on the practical applications of this potentially world-altering technology. It is absolutely essential that publicly powered technology is answerable to public power.

Nathalie Olah is the author of Steal As Much As You Can

Read more: https://www.theguardian.com/commentisfree/2019/dec/02/big-tech-social-remedies-public-ownership-facebook-google

Continue Reading

Ai

Are drone swarms the future of aerial warfare?

Published

on

By

Technology of deploying drones in squadrons is in its infancy, but armed forces are investing millions in its development

As evening fell on Russias Khmeimim airbase in western Syria, the first drones appeared. Then more, until 13 were flashing on radars, speeding towards the airbase and a nearby naval facility.

The explosives-armed aircraft were no trouble for Russian air defences, which shot down seven and jammed the remaining six, according to the countrys defence ministry. But the failed attack in January last year was disturbing to close observers of drone warfare.

It was the first instance of a mass-drone attack and the highest number of drones that I believe weve seen non-state actors use simultaneously in a combat operation, says Paul Scharre, a defence analyst and author who studies the weaponisation of artificial intelligence.

The attempted attacks continued and in September the Russian army said it had downed nearly 60 drones around the Khmeimim base so far this year.

Russian
A Russian general presents what he says are drones that were intercepted near the Khmeimim base. Photograph: Maxime Popov/AFP via Getty Images

For now, military drone use is dominated by lightweight surveillance unmanned aerial vehicles (UAVs) and larger attack UAVs. This situation is unlikely to change in the near future: according to defence experts at the information group Janes, orders for both types of device are expected to increase dramatically in the decade ahead.

But the assaults on Khmeimim, as well as Septembers successful strike on oil facilities in Saudi Arabia, were early flashes of a possible future for aerial warfare: drone swarming.

The technology of swarming drones deployed in squadrons, able to think independently and operate as a pack is in its infancy, but armed forces around the world, including in the UK, are investing millions of pounds in its development.

Smoke
Smoke rises from Saudi Aramcos Abqaiq oil processing facility on 14 September. Photograph: AP

The drones used to attack Khmeimim and the Saudi facilities were likely to have been programmed with the GPS coordinates of their targets and then launched in their direction. Israel is already using hordes of drones to overwhelm Syrian air defences, saturating areas with more targets than anti-aircraft systems can handle.

According to analysts, drone swarms of the future could have the capacity to assess targets, divide up tasks and execute them with limited human interaction.

The real leap forward is swarming where a human says Go accomplish this task and the robots in the swarm communicate amongst each other about how to divvy it up, Scharre says.

A
A test at China Lake, California, shows drone swarms forming an attack orbit. Photograph: US Department of Defence

Analysts predict we might see rudimentary versions of the technology in use within a decade. That might include swarms of drones operating on multiple different frequencies, so they are more resistant to jamming, or swarms that can block or shoot down multiple threats more quickly than the human brain can process.

Two fielders running to catch a ball can [usually] coordinate amongst themselves, Scharre says. But imagine a world where you have 50 fielders and 50 balls. Humans couldnt handle the complexity of that degree of coordination. Robots could handle that with precision.

Advances in swarming technology are mostly classified, though governments have given glimpses of their progress.

In 2016, the US released video of more than 100 micro-drones over a lake in California manoeuvring as a collective organism, sharing one distributed brain for decision-making and adapting to each other like swarms in nature, an air force scientist said.

Play Video
0:54

Footage shows 2016 drone swarm test over lake in California video

In tests last year, the Defense Advanced Research Projects Agency claimed a small squadron of its drones had successfully shared information, allocated jobs and made coordinated tactical decisions against both pre-programmed and pop-up threats.

The US navy has already announced breakthroughs in autonomous boats that could sweep for mines, or serve effectively as bodyguards for larger, manned vessels.

If you look back at the USS Cole bombing that boat was just sitting as an open target at that port in Yemen, says Dan Gettinger, a co-director at the Center for the Study of the Drone at Bard College, referring to the October 2000 attack by two boat-borne al-Qaida suicide bombers that killed 17 American sailors.

If you had a protective shield of unmanned service vehicles, they could intercept that before it happens, he says.

The idea of autonomous, intelligent drones empowered to kill understandably sparks concern. Antnio Guterres, the UN secretary-general, said in a speech last year: The prospect of machines with the discretion and power to take human life is morally repugnant.

In 2017, advocates of a ban against autonomous weapons released a short film, Slaughterbots, depicting a dystopian future where terrorists could unleash swarms of tiny drones capable of identifying and killing specific people.

Some analysts are sceptical of these nightmare scenarios. Drones may one day develop the capacity to carry out targeted killings in swarms. But militaries are not certain to adopt such technology, says Jack Watling, a senior fellow at the Royal United Services Institute.

Their reluctance would be more about expense than ethics. If you think about the logistics of having a lot of sophisticated drones that can pick out individuals, process the data, communicate with each other, navigate a city theres a lot of moving parts to that and its very expensive, Watling says.

More affordable, and therefore more likely to be procured, he says, will be drone swarms that perform relatively simple tasks such as cluttering radar systems to distract and confuse enemy sensors.

Part of what makes drones so attractive is their low cost, Scharre adds. Western military inventories have drastically shrunk in past years, as ships and aircraft have become more sophisticated and too expensive to purchase in large quantities (which, in turn, raises the cost of each vessel or plane).

Drones are a cheap way to boost the sheer size of a force. Western militaries are trying to find ways to add numbers to the equation, to complement these expensive, bespoke aircraft and ships with cheaper systems that can augment them, Scharre says.

Ultimately, he adds, it may be fruitless to try to predict the future of swarming technology from the vantage point of 2019. Imagine someone looking at an airplane in 1912, he says. They might be thinking, This will be useful. But nobody really knows yet what it can do.

Read more: https://www.theguardian.com/news/2019/dec/04/are-drone-swarms-the-future-of-aerial-warfare

Continue Reading

Ai

AWS announces DeepComposer, a machine-learning keyboard for developers

Published

on

By

Today, as AWS re:Invent begins, Amazon announced DeepComposer, a machine learning-driven keyboard aimed at developers.

“AWS DeepComposer is a 32-key, 2-octave keyboard designed for developers to get hands on with Generative AI, with either pretrained models or your own,” AWS’ Julien Simon wrote in a blog post introducing the company’s latest machine learning hardware.

The keyboard is supposed to help developers learn about machine learning in a fun way, and maybe create some music along the way. The area involved in generating creative works in artificial intelligence is called “generative AI.” In other words, it helps you teach machines to generate something creative using “generative adversarial networks.”

“Developers, regardless of their background in ML or music, can get started with Generative Adversarial Networks (GANs). This Generative AI technique pits two different neural networks against each other to produce new and original digital works based on sample inputs. With AWS DeepComposer, you can train and optimize GAN models to create original music,” according to Amazon.

AWS DeepComposer keyboard

Developers can train their own machine learning models or use ones supplied by Amazon to get started. Either way, you create the music based on the model, tweak it in the DeepComposer console on the AWS cloud, then generate your music. If you wish, you can share your machine-generated composition on SoundCloud when you’re done.

This is the third machine learning teaching device from Amazon, joining the DeepLens camera introduced in 2017 and the DeepRacer racing cars introduced last year. It’s worth noting that this is just an announcement. The device isn’t quite ready yet, but Amazon is allowing account holders to sign up for a preview when it is.

Read more: https://techcrunch.com/2019/12/02/aws-announces-deepcomposer-a-machine-learning-keyboard-for-developers/

Continue Reading

Trending