AI algorithms uncannily good at spotting your race from medical scans

Neural networks can correctly guess a person’s race just by looking at their bodily x-rays and researchers have no idea how it can tell.

There are biological features that can give clues to a person’s ethnicity, like the colour of their eyes or skin. But beneath all that, it’s difficult for humans to tell. That’s not the case for AI algorithms, according to a study that’s not yet been peer reviewed.

A team of researchers trained five different models on x-rays of different parts of the body, including chest and hands and then labelled each image according to the patient’s race. The machine learning systems were then tested on how well they could predict someone’s race given just their medical scans.

They were surprisingly accurate. The worst performing was able to predict the right answer 80 per cent of the time, and the best was able to do this 99 per cent, according to the paper.

“We demonstrate that medical AI systems can easily learn to recognise racial identity in medical images, and that this capability is extremely difficult to isolate or mitigate,” the team warns [PDF].

“We strongly recommend that all developers, regulators, and users who are involved with medical image analysis consider the use of deep learning models with extreme caution. In the setting of x-ray and CT imaging data, patient racial identity is readily learnable from the image data alone, generalises to new settings, and may provide a direct mechanism to perpetuate or even worsen the racial disparities that exist in current medical practice.”

Source: AI algorithms uncannily good at spotting your race from medical scans, boffins warn • The Register

Chinese scientists develop world’s strongest glass that’s harder than diamond

Scientists in China have developed the hardest and strongest glassy material known so far that can scratch diamond crystals with ease.

The researchers, including those from Yanshan University in China, noted that the new material – tentatively named AM-III – has “outstanding” mechanical and electronic properties, and could find applications in solar cells due to its “ultra-high” strength and wear resistance.

Analysis of the material, published in the journal National Science Review, revealed that its hardness reached 113 gigapascals (GPa) while natural diamond stone usually scores 50 to 70 on the same test.

[…]

Using fullerenes, which are materials made of hollow football-like arrangements of carbon atoms, the researchers produced different types of glassy materials with varying molecular organisation among which AM-III had the highest order of atoms and molecules.

To achieve this order of molecules, the scientists crushed and blended the fullerenes together, gradually applying intense heat and pressure of about 25 GPa and 1,200 degrees Celsius in an experimental chamber for about 12 hours, spending an equal amount of time cooling the material.

[…]

 

Source: Chinese scientists develop world’s strongest glass that’s as hard as diamond | The Independent

Ancestry.com Gave Itself the Rights to Your Family Photos

The Blackstone-owned genealogy giant Ancestry.com raised a ton of red flags earlier this month with an update to its terms and conditions that give the company a bit more power over your family photos. From here on out, the August 3 update reads, Ancestry can use these pics for any reason, at any time, forever.

[…]

By submitting User Provided Content through any of the Services, you grant Ancestry a perpetual, sublicensable, worldwide, non-revocable, royalty-free license to host, store, copy, publish, distribute, provide access to, create derivative works of, and otherwise use such User Provided Content to the extent and in the form or context we deem appropriate on or through any media or medium and with any technology or devices now known or hereafter developed or discovered. This includes the right for Ancestry to copy, display, and index your User Provided Content. Ancestry will own the indexes it creates.

[…]

The company also noted that it added a helpful clause to clarify that, yes, deleting your documents from Ancestry’s site would also remove any rights Ancestry holds over them. But there’s a catch: if any other Ancestry users copied or saved your content, then Ancestry still holds those rights until these other users delete your documents, too.

[…]

Source: Ancestry.com Gave Itself the Rights to Your Family Photos

Cross-Chain DeFi Site Poly Network Hacked; Hundreds of Millions Potentially Lost

Cross-chain decentralized finance (DeFi) platform Poly Network was attacked on Tuesday, with the alleged hacker draining roughly $600 million in crypto.

Poly Network, a protocol launched by the founder of Chinese blockchain project Neo, operates on the Binance Smart Chain, Ethereum and Polygon blockchains. Tuesday’s attack struck each chain consecutively, with the Poly team identifying three addresses where stolen assets were transferred.

At the time that Poly tweeted news of the attack, the three addresses collectively held more than $600 million in different cryptocurrencies, including USDC, wrapped bitcoin (WBTC, -1.45%), wrapped ether (ETH, -0.7%) and shiba inu (SHIB), blockchain scanning platforms show.

[…]

About one hour after Poly announced the hack on Twitter, the hacker tried to move assets including USDT through the Ethereum address into liquidity pool Curve.fi, records show. The transaction was rejected.

Meanwhile, close to $100 million has been moved out of the Binance Smart Chain address in the past 30 minutes and deposited into liquidity pool Ellipsis Finance.

[…]

BlockSec, a China-based blockchain security firm, said in an initial attack analysis report that the hack may be triggered by the leak of a private key that was used to sign the cross-chain message.

But it also added that another possible reason is a potential bug during Poly’s singing process that may have been “abused” to sign the message.

According to another China-based blockchain security firm, Slowmist, the attackers’ original funds were in monero (XMR, -2.9%), a privacy-centric cryptocurrency, and were then exchanged for BNB, ETH, MATIC (+0.86%) and a few other tokens.

The attackers then initiated the attacks on Ethereum, BSC and Polygon blockchains. The finding was supported by Slowmist’s partners, including China-based exchange Hoo.

“Based on the flows of the funds and multiple fingerprint information, it is likely a long-planned, organized, and well-prepared attack,” Slowmist wrote.

[…]

The Poly Network incident shows how nascent cross-chain protocols are particularly vulnerable to attacks. In July, cross-chain liquidity protocol Thorchain suffered two exploits in two weeks. Rari Capital, another cross-chain DeFi protocol, was hit by an attack in May, losing funds worth nearly $11 million in ETH.

[…]

Source: Cross-Chain DeFi Site Poly Network Hacked; Hundreds of Millions Potentially Lost – CoinDesk