Peanut allergy cured for 4 years in majority of children in immunotherapy trial

A small clinical trial conducted at the Murdoch Children’s Research Institute has led to two-thirds of children treated with an experimental immunotherapy treatment being cured of their allergy. Importantly, this desensitisation to peanuts persisted for up to four years after treatment.
[…]
Forty-eight children were enrolled in the PPOIT trial and were randomly given either a combination of the probiotic Lactobacillus rhamnosus with peanut protein in increasing amounts, or a placebo, once daily for 18 months.

At the end of the original trial in 2013, 82% of children who received the immunotherapy treatment were deemed tolerant to peanuts compared with just 4% in the placebo group.

Four years later, the majority of the children who gained initial tolerance were still eating peanuts as part of their normal diet and 70% passed a further challenge test to confirm long-term tolerance.

Source: Peanut allergy cured in majority of children in immunotherapy trial

USA: those massive terms &c you never read are legally binding: and can stop you from using the legal system to sue! (Victory for Uber!)

You may never read those lengthy terms and conditions attached to every digital download or app but, in America at least, they are legally binding. Sorry.

That’s the conclusion of a panel of appeal judges earlier this week when shining beacon of corporate responsibility Uber insisted its users had agreed not to sue the company somewhere in its long list of lengthy legal locutions.

On Thursday, the US Second Court of Appeals decided [PDF] that when customers installed Uber’s ride-hailing app and agreed to the terms and conditions – even though virtually none of them actually read the details – they were obliged to go through arbitration if they had a dispute with the company.

The case was very closely watched by technology companies for obvious reasons – if the court ruled differently it could have opened them up to a wave of potential liability and public scrutiny.

As it stands, the arbitration requirement will hold: a situation that enables many companies to keep embarrassing cock-ups and business practices under wraps since unhappy consumers are obliged to go through the process privately and details are not made public.

Source: Sorry, but those huge walls of terms and conditions you never read are legally binding

Absolute legal lunacy!

Boeing draws a plane in the sky with flight path

In a test flight, Boeing took the thing where you draw using your GPS path to a whole different level. They drew the outline of a plane that spanned the latitude of the conterminous United States.

Source: Boeing draws a plane in the sky with flight path | FlowingData

Kino: Kinetic, Robot jewels that crawl up and down you

https://www.media.mit.edu/projects/kino-kinetic-wearable/overview/

UK Home Secretary calls people who use encryption not ‘real’ and Daesh sympathisers

In an article in the Daily Telegraph timed to coincide with Rudd’s appearance at a closed event in San Francisco, Rudd argued: “Real people often prefer ease of use and a multitude of features to perfect, unbreakable security.”

She continued: “Who uses WhatsApp because it is end-to-end encrypted, rather than because it is an incredibly user-friendly and cheap way of staying in touch with friends and family? Companies are constantly making trade-offs between security and ‘usability,’ and it is here where our experts believe opportunities may lie.”

The reference to “real people” struck a nerve with a host of security experts, sysadmins, privacy advocates and tech-savvy consumers who took to Twitter to point out that they were real people, and not ISIS sympathizers – as Rudd implied in her piece. Rudd essentially declared that people who use strong encryption are not normal, not real people, which is a rather dangerous sentiment.

Source: ‘Real’ people want govts to spy on them, argues UK Home Secretary

What the actual fuck?

US Congress dreams of IoT and gets it right! Except it won’t protect consumers, only gov.

The Internet of Things Cybersecurity Improvement Act would require that IoT devices purchased by the American government must not have any known security vulnerabilities, must have the ability to be patched, and may not have hardcoded passwords built in. It mandates that every government department inventory all IoT devices on their networks.
[…]
The bill also directs Homeland Security to come up with a vulnerability disclosure program so that departments can get patched and updated. Another requirement says the Office of Management and Budget must come up with reasonable standards as to what IoT security should actually entail.
[…]
A key element of the proposed legislation is that it would make it legal for security researchers to tear these devices apart and search for security bugs. Currently a broad interpretation of the Digital Millennium Copyright Act means that a company could prosecute a researcher who looks into the firmware for breaking the terms and conditions of its use.

Source: No vulns. No hardwired passwords. Patchable. Congress dreams of IoT: Impossible Online Tech

US Secret Service agent Bridges broadcast Bitstamp Bitcoins to BTC-E besides Silk Road heist

Shaun Bridges, who is already serving a six-year sentence for nicking Bitcoins from the underground souk, pleaded guilty on Tuesday to stealing a further 1,600 Bitcoin (worth $359,005 at the time and approximately $6.6m today) during a separate investigation.

According to court documents [PDF] Bridges, 35, was probing European Bitcoin trading firm Bitstamp, which led to the US government seizing 1,606,6488 BTC in November 2014. These were transferred into a digital wallet that only Bridges had the access code for.

In March 2015, while under investigation for the Silk Road thefts, Bridges resigned from the Secret Service and in June pleaded guilty to money laundering and obstruction charges. A month later, while still free and awaiting sentencing, he took the Bitcoins seized from Bitstamp and moved them into an account run by the BTC-E exchange.

Source: Disgraced US Secret Service agent coughs to second Bitcoin heist

OpenAI bot bursts into the ring, humiliates top Dota 2 pro gamer in ‘scary’ one-on-one bout

In a shock move on Friday evening, the software agent squared up to top Dota 2 pro gamer Dendi, a Ukrainian 27-year-old, at the Dota 2 world championships dubbed The International.

The OpenAI agent beat Dendi in less than 10 minutes in the first round, and trounced him again in a second round, securing victory in a best-of-three match. “This guy is scary,” a shocked Dendi told the huge crowd watching the battle at the event. Musk was jubilant.
[…]
According to OpenAI, its machine-learning bot was also able to pwn two other top human players earlier this week: SumaiL and Arteezy. Although it’s an impressive breakthrough, it’s important to note this popular strategy game is usually played not one-v-one but as a five-versus-five team game – a rather difficult environment for bots to handle.

Source: OpenAI bot bursts into the ring, humiliates top Dota 2 pro gamer in ‘scary’ one-on-one bout

rechargable safe battery

Through Ionic Materials’ invention of a novel solid polymer electrolyte material that conducts ions at room temperature, we are on the verge of revolutionizing battery technology. A truly solid state battery is now possible. Significant improvements in battery safety, performance and cost are achievable with ionic conductivities that exceed those of traditional liquid systems over a wide range of temperatures.

Source: the solution | Ionic Materials

MIT Real time automatic image retouching on your phone

System can apply a range of styles in real-time, so that the viewfinder displays the enhanced image.
[…]
at Siggraph, the premier digital graphics conference, researchers from MIT’s Computer Science and Artificial Intelligence Laboratory and Google are presenting a new system that can automatically retouch images in the style of a professional photographer. It’s so energy-efficient, however, that it can run on a cellphone, and it’s so fast that it can display retouched images in real-time, so that the photographer can see the final version of the image while still framing the shot.

The same system can also speed up existing image-processing algorithms. In tests involving a new Google algorithm for producing high-dynamic-range images, which capture subtleties of color lost in standard digital images, the new system produced results that were visually indistinguishable from those of the algorithm in about one-tenth the time — again, fast enough for real-time display.

The system is a machine-learning system, meaning that it learns to perform tasks by analyzing training data; in this case, for each new task it learned, it was trained on thousands of pairs of images, raw and retouched.

Source: Automatic image retouching on your phone

70% of Windows 10 users haven’t turned of privacy invasion

Microsoft claims seven out of ten Windows 10 users are happy with Redmond gulping loads of telemetry from their computers – which isn’t that astounding when you realize it’s a default option.

In other words, 30 per cent of people have found the switch to turn it off, and the rest haven’t, don’t realize it’s there, or are genuinely OK with the data collection.
[…]
Essentially, if you’re on Home or Pro, you can’t tell your OS to not phone home. And, sure, this information – from lists of hardware and apps installed to pen gestures – is useful to Microsoft employees debugging code that’s running in the field. But we’re all adults here, and some folks would like the option to not have any information leaving their systems.

Source: 70% of Windows 10 users are totally happy with our big telemetry slurp, beams Microsoft

Nice spin, to say people “choose” the default option, when it isn’t a choice people actually can make!

This is why I am leaving Windows for what it is and moving to Linux Mint.

Disney sued for allegedly spying on children through 42 gaming apps

A federal class action lawsuit filed last week in California alleges that the Walt Disney Company is violating privacy protection laws by collecting children’s personal information from 42 of its apps and sharing the data with advertisers without parental consent.

The lawsuit targets Disney and three software companies — Upsight, Unity, and Kochava — alleging that the companies created mobile apps aimed at children that contained embedded software to track, collect, and then export their personal information along with information about their online behavior. The plaintiff, a San Francisco woman named Amanda Rushing, says she was unaware that information about her child, “L.L.,” was collected while playing mobile game Disney Princess Palace Pets, and that data was then sold to third parties for ad targeting.

The Verge

DNA Testing Data Is Disturbingly Vulnerable to Hackers

In a new study that will be presented next week at the 26th USENIX Security Symposium in Vancouver, University of Washington researchers analyzed the security practices of common, open-source DNA processing programs and found that they were, in general, lacking. That means all that super-sensitive information those programs are processing is potentially vulnerable to hackers. If you think social security fraud is bad, imagine someone hacking your genetic code.

“You can imagine someone altering the DNA at a crime scene, or making it unreadable. Or an attacker stealing data or modifying it in a certain way to make it seem like someone has a disease someone doesn’t actually have,” Peter Ney, a co-author of the peer-reviewed study and Ph.D. student at the school’s Computer Security and Privacy Research Lab, told Gizmodo

Source: DNA Testing Data Is Disturbingly Vulnerable to Hackers

Emma – Plagiarism detecting AI

Emma is a self-learning technology. She analyses and understands the way people write.

Emma uses machine learning and artificial intelligence to study the innards of each author’s writing style and attributes authorship on their basis.

Emma Identity

Artificially intelligent painters invent new styles of art

The team – which also included researchers at Rutgers University in New Jersey and Facebook’s AI lab in California – modified a type of algorithm known as a generative adversarial network (GAN), in which two neural nets play off against each other to get better and better results. One creates a solution, the other judges it – and the algorithm loops back and forth until the desired result is reached.

In the art AI, one of these roles is played by a generator network, which creates images. The other is played by a discriminator network, which was trained on 81,500 paintings to tell the difference between images we would class as artworks and those we wouldn’t – such as a photo or diagram, say.

The discriminator was also trained to distinguish different styles of art, such as rococo or cubism.
Art with a twist

The clever twist is that the generator is primed to produce an image that the discriminator recognises as art, but which does not fall into any of the existing styles.

“You want to have something really creative and striking – but at the same time not go too far and make something that isn’t aesthetically pleasing,” says team member Ahmed Elgammal at Rutgers University.

Once the AI had produced a series of images, members of the public were asked to judge them alongside paintings by people in an online survey, without knowing which were the AI’s work. Participants answered questions about how complex or novel they felt each image was, and whether it inspired them or elevated their mood. To the researchers’ surprise, images produced by their AI scored slightly higher in many cases than those by humans.

New Scientist

With a single wiretap order, US authorities listened in on 3.3 million phone calls

US authorities intercepted and recorded millions of phone calls last year under a single wiretap order, authorized as part of a narcotics investigation.

The wiretap order authorized an unknown government agency to carry out real-time intercepts of 3.29 million cell phone conversations over a two-month period at some point during 2016, after the order was applied for in late 2015.

The order was signed to help authorities track 26 individuals suspected of involvement with illegal drug and narcotic-related activities in Pennsylvania.

The wiretap cost the authorities $335,000 to conduct and led to a dozen arrests.

But the authorities noted that the surveillance effort led to no incriminating intercepts, and none of the handful of those arrested have been brought to trial or convicted.

Solar panel waste creates 300 times more toxic waste per unit of energy than do nuclear power plants.

Only Europe requires solar panel makers to collect and dispose of solar waste at the end of their lives.

All of which raises the question: just how big of a problem is solar waste?

Environmental Progress investigated the problem to see how the problem compared to the much more high-profile issue of nuclear waste.

We found:

Solar panels create 300 times more toxic waste per unit of energy than do nuclear power plants.

If solar and nuclear produce the same amount of electricity over the next 25 years that nuclear produced in 2016, and the wastes are stacked on football fields, the nuclear waste would reach the height of the Leaning Tower of Pisa (52 meters), while the solar waste would reach the height of two Mt. Everests (16 km).

In countries like China, India, and Ghana, communities living near e-waste dumps often burn the waste in order to salvage the valuable copper wires for resale. Since this process requires burning off the plastic, the resulting smoke contains toxic fumes that are carcinogenic and teratogenic (birth defect-causing) when inhaled.

Environmental progress

Time for other countries to start thinking about this, if it’s true?

Draw Together with a Neural Network

We made an interactive web experiment that lets you draw together with a recurrent neural network model called sketch-rnn. We taught this neural net to draw by training it on millions of doodles collected from the Quick, Draw! game. Once you start drawing an object, sketch-rnn will come up with many possible ways to continue drawing this object based on where you left off. Try the first demo.

tensorflow.org

California generates 1/2 the US solar energy, has to pay neighbouring states to take the energy

California is the poster child for solar energy: in 2016, 13% of the state’s power came from solar sources. According to the Solar Energy Industries Association, California is in the lead for the cumulative amount of solar electric capacity installed in 2016.

In fact, the California is generating so much solar energy that it is resorting to paying other states to take the excess electricity in order to prevent overloading power lines. According to the Los Angeles Times, Arizona residents have already saved millions in 2017 thanks to California’s contribution.

The state, which produced little to no solar energy just 15 years ago, has made strides — it single-handedly has nearly half of the country’s solar electricity generating capacity. According to the U.S. Energy Information Administration, California reached a milestone: for a few hours, more than half the state’s power needs were sourced from solar energy. This put wholesale energy prices in the negative.

mic.com

3 Etherium heists in as many weeks: $7m, $32m and $85m!

Hacker Allegedly Steals $7.4 Million in Ethereum with Incredibly Simple Trick

Someone tricked would be investors during an ethereum ICO into sending their cryptocurrency to the wrong address.

A hacker has allegedly just stolen around $7.4 million dollars worth of ether, the cryptocurrency that underpins the app platform ethereum, by tricking victims into sending money to the wrong address during an Initial Coin Offering, or ICO. This is according to a company called Coindash that says its investors were sending their funds to a hacker.

Hacker Uses Parity Wallet Vulnerability to Steal $30 Million Worth of Ethereum

An unknown hacker has used a vulnerability in an Ethereum wallet client to steal over 153,000 Ether, worth over $30 million dollars.

The hack was possible due to a flaw in the Parity Ethereum client. The vulnerability allowed the hacker to exfiltrate funds from multi-sig wallets created with Parity clients 1.5 and later. Parity 1.5 was released on January 19, 2017.

Multi-sig wallets are Ethereum accounts over which multiple persons have control with their own keys. Multi-sig accounts allow owners to move funds only when a majority of owners sign a transaction with their key.

These hackers stole $85 million in ether to save it from *the real crooks* (or so they say)

The clock was ticking. Thieves stole $32 million worth of ether out of a popular Ethereum wallet, and with every passing minute the potential for additional losses grew.

And so the White Hat Group stepped in.

Like something out of a weird cryptocurrency reboot of National Treasure, the unidentified WHG hackers decided to steal the remaining ether before the crooks could. All $85 million of it.

Or so they say.

The claim was posted to Reddit on July 19, and details a plan to return the funds to their rightful owners. Here’s how the poster, jbaylina, says it went down:

“The White Hat Group were made aware of a vulnerability in a specific version of a commonly used multisig contract,” explained the post, referring to a vulnerability in the popular Ethereum wallet Parity that was successfully exploited by unknown thieves. “This vulnerability was trivial to execute, so they took the necessary action to drain every vulnerable multisig they could find as quickly as possible. Thank you to the greater Ethereum Community that helped finding these vulnerable contracts.”

Intel Launches Movidius Neural Compute Stick: Deep Learning and AI on a $79 USB Stick

Meanwhile, the on-chip memory has increased from 1 GB on the Fathom NCS to 4 GB LPDDR3 on the Movidius NCS, in order to facilitate larger and denser neural networks. And to cap it all off, Movidius has been able to reduce the MSRP to $79 – citing Intel’s “manufacturing and design expertise” – lowering the cost of entry even more.

Like other players in the edge inference market, Movidius is looking to promote and capitalize on the need for low-power but capable inference processors for stand-alone devices. That means targeting use cases where the latency of going to a server would be too great, a high-performance CPU too power hungry, or where privacy is a greater concern. In which case, the NCS and the underlying Myriad 2 VPU are Intel’s primary products for device manufacturers and software developers.

Source: Intel Launches Movidius Neural Compute Stick: Deep Learning and AI on a $79 USB Stick

Swedish government leak: clueless agency moved all citizens data + military secrets to “The Cloud” in clear text and to people without security clearances in many countries

Sweden’s Transport Agency moved all of its data to “the cloud”, apparently unaware that there is no cloud, only somebody else’s computer. In doing so, it exposed and leaked every conceivable top secret database: fighter pilots, SEAL team operators, police suspects, people under witness relocation. Names, photos, and home addresses: the list is just getting started. The responsible director has been found guilty in criminal court of the whole affair, and sentenced to the harshest sentence ever seen in Swedish government: she was docked half a month’s paycheck.
[…]
Last March, the entire register of vehicles was sent to marketers subscribing to it. This is normal in itself, as the vehicle register is public information, and therefore subject to Freedom-of-Information excerpts. What was not normal were two things: first, that people in the witness protection program and similar programs were included in the register distributed outside the Agency, and second, when this fatal mistake was discovered, a new version without the sensitive identities was not distributed with instructions to destroy the old copy. Instead, the sensitive identities were pointed out and named in a second distribution with a request for all subscribers to remove these records themselves. This took place in open cleartext e-mail.
[…]
The weight capacity of all roads and bridges (which is crucial for warfare, and says a lot about what roads are intended to be used as wartime airfields);

Names, photos, and home addresses of fighter pilots in the Air Force;

Names, photos, and home addresses of everybody and anybody in a police register, all of which are classified;

Names, photos, and home addresses of all operators in the military’s most secret units – equivalent to the SAS or SEAL teams;

Names, photos, and home addresses of everybody in a witness relocation program or who has been given protected identity for other reasons;

Type, model, weight, and any defects of any and all government and military vehicles, including their operator, which says a ton about the structure of military support units;

[…]
All of this was not just outside the proper agencies, but outside the European Union, in the hands of people who had absolutely no security clearance. All of this data can be expected to have been permanently exposed.

Source: Worst government leak: clueless agency moved everything to “The Cloud”

Just completely wow!

Lenovo Folio: 5.5″ phone that unfolds into 8″ tablet seamlessly

At the third annual Lenovo Tech World last week, the Chinese tech giant wowed attendees with the Lenovo Folio, a tablet with a screen that folds in half into a phone.

Before you start getting too excited, you should know that the Folio is a concept device, which means it may not be released as a consumer product anytime soon. Even so, that doesn’t make the device any less impressive.

The tablet has a 7.8-inch screen with 1,920 x 1,440 resolution, a Qualcomm Snapdragon 800 processor, and runs Android 7.0 Nougat. It’s not exactly peak performance in 2017, but that’s not why’d you want this thing — you’d want it for the bendable screen.

When folded, the tablet shrinks down into a 5.5-inch phone that could fit into your pocket. As you can see in the demo videos above and below, the display folds neatly in half with pixels filling all the space where a hinge would normally. The UI automatically adjusts to work as if there are two displays. It’s pretty bananas!

Mashable

8″ is the perfect tablet size IMHO – a real shame nobody makes them anymore either…

AI quickly cooks malware that AV software can’t spot

Hyrum Anderson, technical director of data science at security shop Endgame, showed off research that his company had done in adapting Elon Musk’s OpenAI framework to the task of creating malware that security engines can’t spot.

The system basically learns how to tweak malicious binaries so that they can slip past antivirus tools and continue to work once unpacked and executed. Changing small sequences of bytes can fool AV engines, even ones that are also powered by artificial intelligence, he said. Anderson cited research by Google and others to show how changing just a few pixels in an image can cause classification software to mistake a bus for an ostrich.

“All machine learning models have blind spots,” he said. “Depending on how much knowledge a hacker has they can be convenient to exploit.”

So the team built a fairly simple mechanism to develop weaponised code by making very small changes to malware and firing these variants at an antivirus file scanner. By monitoring the response from the engine they were able to make lots of tiny tweaks that proved very effective at crafting software nasties that could evade security sensors.

The malware-tweaking machine-learning software was trained over 15 hours and 100,000 iterations, and then lobbed some samples at an antivirus classifier. The attacking code was able to get 16 per cent of its customized samples past the security system’s defenses, we’re told.

This software-generation software will be online at the firm’s Github page and Anderson encouraged people to give it a try. No doubt security firms will also be taking a long look at how this affects their products in the future

Source: AI quickly cooks malware that AV software can’t spot

It is easy to expose users’ secret web habits, if you have access to cheap clickstream data

Two German researchers say they have exposed the porn-browsing habits of a judge, a cyber-crime investigation and the drug preferences of a politician.

The pair obtained huge amounts of information about the browsing habits of three million German citizens from companies that gather “clickstreams”.

These are detailed records of everywhere that people go online.

The researchers argue such data – which some firms scoop up and use to target ads – should be protected.
[…]
The pair found that 95% of the data they obtained came from 10 popular browser extensions.
[…]
The public information included links people shared via Twitter, YouTube videos they reported watching, news articles they passed on via social media or when they posted online photos of items they bought or places they visited.

In many cases, he said, it was even easier to de-anonymise because the clickstreams contained links to people’s personal social media admin pages which directly revealed their identity.

Source: It is easy to expose users’ secret web habits, say researchers – BBC News

 
Skip to toolbar