New property of light discovered, plus recently discovered properties you probably didn’t know about

Scientists have long known about such properties of light as wavelength. More recently, researchers have found that light can also be twisted, a property called . Beams with highly structured angular momentum are said to have orbital angular momentum (OAM), and are called . They appear as a helix surrounding a common center, and when they strike a flat surface, they appear as doughnut-shaped. In this new effort, the researchers were working with OAM beams when they found the light behaving in a way that had never been seen before.

The experiments involved firing two lasers at a cloud of argon gas—doing so forced the beams to overlap, and they joined and were emitted as a single beam from the other side of the argon cloud. The result was a type of vortex beam. The researchers then wondered what would happen if the lasers had different orbital angular momentum and if they were slightly out of sync. This resulted in a beam that looked like a corkscrew with a gradually changing twist. And when the beam struck a , it looked like a crescent moon. The researchers noted that looked at another way, a at the front of the beam was orbiting around its center more slowly than a photon at the back of the . The researchers promptly dubbed the new property self-torque—and not only is it a newly discovered property of light, it is also one that has never even been predicted.

00:00
00:00
A new property of light beams, the self-torque of light, which is associated to a temporal variation of the orbital angular momentum. Extreme-ultraviolet ultrafast pulses with self-torque are generated through high harmonic generation. Credit: JILA (USA) Rebecca Jacobson, Servicio de Produccion e Innovacion Digital – Universidad de Salamanca (Spain)

The researchers suggest that it should be possible to use their technique to modulate the of light in ways very similar to modulating frequencies in communications equipment. This could lead to the development of novel devices that make use of manipulating extremely tiny materials.

Source: New property of light discovered

Researchers teleport information within a diamond

Researchers from the Yokohama National University have teleported quantum information securely within the confines of a diamond. The study has big implications for quantum information technology—the future of sharing and storing sensitive information. The researchers published their results on June 28, 2019, in Communications Physics.

“Quantum teleportation permits the transfer of into an otherwise inaccessible space,” said Hideo Kosaka, a professor of engineering at Yokohama National University and an author on the study. “It also permits the transfer of information into a quantum memory without revealing or destroying the stored quantum information.”

The inaccessible space, in this case, consisted of in diamond. Made of linked, yet individually contained, carbon atoms, a diamond holds the perfect conditions for .

A carbon atom holds six protons and six neutrons in its nucleus, surrounded by six spinning electrons. As the atoms bond into a diamond, they form a notably strong lattice. However, can have complex defects, such as when a nitrogen atom exists in one of two adjacent vacancies where carbon atoms should be. This defect is called a nitrogen vacancy center.

Surrounded by carbon atoms, the nucleus structure of the creates what Kosaka calls a nanomagnet.

To manipulate an electron and a carbon isotope in the vacancy, Kosaka and the team attached a wire about a quarter the width of a human hair to the surface of a diamond. They applied a microwave and a radio wave to the wire to build an oscillating magnetic field around the diamond. They shaped the microwave to create the optimal, controlled conditions for the transfer of quantum information within the diamond.

Kosaka then used the nitrogen nanomagnet to anchor an electron. Using the microwave and radio waves, Kosaka forced the to entangle with a carbon nuclear spin—the angular momentum of the electron and the nucleus of a carbon atom. The electron spin breaks down under a created by the nanomagnet, making it susceptible to entanglement. Once the two pieces are entangled, meaning their physical characteristics are so intertwined they cannot be described individually, a photon that holds quantum information is introduced, and the electron absorbs the photon. The absorption allows the polarization state of the photon to be transferred into the carbon, which is mediated by the entangled electron, demonstrating a teleportation of information at the quantum level.

“The success of the photon storage in the other node establishes the entanglement between two adjacent nodes,” Kosaka said. Called quantum repeaters, the process can take individual chunks of information from node to node, across the quantum field.

“Our ultimate goal is to realize scalable quantum repeaters for long-haul quantum communications and distributed quantum computers for large-scale quantum computation and metrology,” Kosaka said.

Source: Researchers teleport information within a diamond

That this AI can simulate universes in 30ms is not the scary part. It’s that its creators don’t know why it works so well

The accuracy of the neural network is judged by how similar its outputs were to the ones created by two more traditional N-body simulation systems, FastPM and 2LPT, when all three are given the same inputs. When D3M was tasked with producing 1,000 simulations from 1,000 sets of input data, it had a relative error of 2.8 per cent compared to FastPM, and a 9.3 per cent compared to 2LPT for the same inputs. That’s not too bad, considering it takes the model just 30 milliseconds to crank out a simulation. Not only does that save time, but it’s also cheaper too since less compute power is needed.

To their surprise, the researchers also noticed that D3M seemed to be able to produce simulations of the universe from conditions that weren’t specifically included in the training data. During inference tests, the team tweaked input variables such as the amount of dark matter in the virtual universes, and the model still managed to spit out accurate simulations despite not being specifically trained for these changes.

“It’s like teaching image recognition software with lots of pictures of cats and dogs, but then it’s able to recognize elephants,” said Shirley Ho, first author of the paper and a group leader at the Flatiron Institute. “Nobody knows how it does this, and it’s a great mystery to be solved.

“We can be an interesting playground for a machine learner to use to see why this model extrapolates so well, why it extrapolates to elephants instead of just recognizing cats and dogs. It’s a two-way street between science and deep learning.”

The source code for the neural networks can be found here.

Source: That this AI can simulate universes in 30ms is not the scary part. It’s that its creators don’t know why it works so well • The Register

EU should ban AI-powered citizen scoring and mass surveillance, say experts

A group of policy experts assembled by the EU has recommended that it ban the use of AI for mass surveillance and mass “scoring of individuals”; a practice that potentially involves collecting varied data about citizens — everything from criminal records to their behavior on social media — and then using it to assess their moral or ethical integrity.

The recommendations are part of the EU’s ongoing efforts to establish itself as a leader in so-called “ethical AI.” Earlier this year, it released its first guidelines on the topic, stating that AI in the EU should be deployed in a trustworthy and “human-centric” manner.

The new report offers more specific recommendations. These include identifying areas of AI research that require funding; encouraging the EU to incorporate AI training into schools and universities; and suggesting new methods to monitor the impact of AI. However, the paper is only a set of recommendations at this point, and not a blueprint for legislation.

Notably, the suggestions that the EU should ban AI-enabled mass scoring and limit mass surveillance are some of the report’s relatively few concrete recommendations. (Often, the report’s authors simply suggest that further investigation is needed in this or that area.)

Source: EU should ban AI-powered citizen scoring and mass surveillance, say experts – The Verge

Google’s new reCaptcha forces page admins to put it on EVERY page so Google can track you everywhere

According to tech statistics website Built With, more than 650,000 websites are already using reCaptcha v3; overall, there are at least 4.5 million websites use reCaptcha, including 25% of the top 10,000 sites. Google is also now testing an enterprise version of reCaptcha v3, where Google creates a customized reCaptcha for enterprises that are looking for more granular data about users’ risk levels to protect their site algorithms from malicious users and bots.

But this new, risk-score based system comes with a serious trade-off: users’ privacy.

According to two security researchers who’ve studied reCaptcha, one of the ways that Google determines whether you’re a malicious user or not is whether you already have a Google cookie installed on your browser. It’s the same cookie that allows you to open new tabs in your browser and not have to re-log in to your Google account every time. But according to Mohamed Akrout, a computer science PhD student at the University of Toronto who has studied reCaptcha, it appears that Google is also using its cookies to determine whether someone is a human in reCaptcha v3 tests. Akrout wrote in an April paper about how reCaptcha v3 simulations that ran on a browser with a connected Google account received lower risk scores than browsers without a connected Google account. “If you have a Google account it’s more likely you are human,” he says. Google did not respond to questions about the role that Google cookies play in reCaptcha.

With reCaptcha v3, technology consultant Marcos Perona and Akrout’s tests both found that their reCaptcha scores were always low risk when they visited a test website on a browser where they were already logged into a Google account. Alternatively, if they went to the test website from a private browser like Tor or a VPN, their scores were high risk.

To make this risk-score system work accurately, website administrators are supposed to embed reCaptcha v3 code on all of the pages of their website, not just on forms or log-in pages. Then, reCaptcha learns over time how their website’s users typically act, helping the machine learning algorithm underlying it to generate more accurate risk scores. Because reCaptcha v3 is likely to be on every page of a website,  if you’re signed into your Google account there’s a chance Google is getting data about every single webpage you go to that is embedded with reCaptcha v3—and there many be no visual indication on the site that it’s happening, beyond a small reCaptcha logo hidden in the corner.

Source: Google’s new reCaptcha has a dark side

8 of worlds top tech companies pwned for years by China

Eight of the world’s biggest technology service providers were hacked by Chinese cyber spies in an elaborate and years-long invasion, Reuters found. The invasion exploited weaknesses in those companies, their customers, and the Western system of technological defense.

[…]

The hacking campaign, known as “Cloud Hopper,” was the subject of a U.S. indictment in December that accused two Chinese nationals of identity theft and fraud. Prosecutors described an elaborate operation that victimized multiple Western companies but stopped short of naming them. A Reuters report at the time identified two: Hewlett Packard Enterprise and IBM.

Yet the campaign ensnared at least six more major technology firms, touching five of the world’s 10 biggest tech service providers.

Also compromised by Cloud Hopper, Reuters has found: Fujitsu, Tata Consultancy Services, NTT Data, Dimension Data, Computer Sciences Corporation and DXC Technology. HPE spun-off its services arm in a merger with Computer Sciences Corporation in 2017 to create DXC.

Waves of hacking victims emanate from those six plus HPE and IBM: their clients. Ericsson, which competes with Chinese firms in the strategically critical mobile telecoms business, is one. Others include travel reservation system Sabre, the American leader in managing plane bookings, and the largest shipbuilder for the U.S. Navy, Huntington Ingalls Industries, which builds America’s nuclear submarines at a Virginia shipyard.

“This was the theft of industrial or commercial secrets for the purpose of advancing an economy,” said former Australian National Cyber Security Adviser Alastair MacGibbon. “The lifeblood of a company.”

[…]

The corporate and government response to the attacks was undermined as service providers withheld information from hacked clients, out of concern over legal liability and bad publicity, records and interviews show. That failure, intelligence officials say, calls into question Western institutions’ ability to share information in the way needed to defend against elaborate cyber invasions. Even now, many victims may not be aware they were hit.

The campaign also highlights the security vulnerabilities inherent in cloud computing, an increasingly popular practice in which companies contract with outside vendors for remote computer services and data storage.

[…]

For years, the company’s predecessor, technology giant Hewlett Packard, didn’t even know it had been hacked. It first found malicious code stored on a company server in 2012. The company called in outside experts, who found infections dating to at least January 2010.

Hewlett Packard security staff fought back, tracking the intruders, shoring up defenses and executing a carefully planned expulsion to simultaneously knock out all of the hackers’ known footholds. But the attackers returned, beginning a cycle that continued for at least five years.

The intruders stayed a step ahead. They would grab reams of data before planned eviction efforts by HP engineers. Repeatedly, they took whole directories of credentials, a brazen act netting them the ability to impersonate hundreds of employees.

The hackers knew exactly where to retrieve the most sensitive data and littered their code with expletives and taunts. One hacking tool contained the message “FUCK ANY AV” – referencing their victims’ reliance on anti-virus software. The name of a malicious domain used in the wider campaign appeared to mock U.S. intelligence: “nsa.mefound.com”

Then things got worse, documents show.

After a 2015 tip-off from the U.S. Federal Bureau of Investigation about infected computers communicating with an external server, HPE combined three probes it had underway into one effort called Tripleplay. Up to 122 HPE-managed systems and 102 systems designated to be spun out into the new DXC operation had been compromised, a late 2016 presentation to executives showed.

[…]

According to Western officials, the attackers were multiple Chinese government-backed hacking groups. The most feared was known as APT10 and directed by the Ministry of State Security, U.S. prosecutors say. National security experts say the Chinese intelligence service is comparable to the U.S. Central Intelligence Agency, capable of pursuing both electronic and human spying operations.

[…]

It’s impossible to say how many companies were breached through the service provider that originated as part of Hewlett Packard, then became Hewlett Packard Enterprise and is now known as DXC.

[…]

HP management only grudgingly allowed its own defenders the investigation access they needed and cautioned against telling Sabre everything, the former employees said. “Limiting knowledge to the customer was key,” one said. “It was incredibly frustrating. We had all these skills and capabilities to bring to bear, and we were just not allowed to do that.”

[…]

The threat also reached into the U.S. defense industry.

In early 2017, HPE analysts saw evidence that Huntington Ingalls Industries, a significant client and the largest U.S. military shipbuilder, had been penetrated by the Chinese hackers, two sources said. Computer systems owned by a subsidiary of Huntington Ingalls were connecting to a foreign server controlled by APT10.

During a private briefing with HPE staff, Huntington Ingalls executives voiced concern the hackers could have accessed data from its biggest operation, the Newport News, Va., shipyard where it builds nuclear-powered submarines, said a person familiar with the discussions. It’s not clear whether any data was stolen.

[…]

Like many Cloud Hopper victims, Ericsson could not always tell what data was being targeted. Sometimes, the attackers appeared to seek out project management information, such as schedules and timeframes. Another time they went after product manuals, some of which were already publicly available.

[…]

much of Cloud Hopper’s activity has been deliberately kept from public view, often at the urging of corporate victims.

In an effort to keep information under wraps, security staff at the affected managed service providers were often barred from speaking even to other employees not specifically added to the inquiries.

In 2016, HPE’s office of general counsel for global functions issued a memo about an investigation codenamed White Wolf. “Preserving confidentiality of this project and associated activity is critical,” the memo warned, stating without elaboration that the effort “is a sensitive matter.” Outside the project, it said, “do not share any information about White Wolf, its effect on HPE, or the activities HPE is taking.”

The secrecy was not unique to HPE. Even when the government alerted technology service providers, the companies would not always pass on warnings to clients, Jeanette Manfra, a senior cybersecurity official with the U.S. Department of Homeland Security, told Reuters.

Source: Stealing Clouds