Google has unveiled an Artificial Intelligence (AI)-based web app that will create a poem portrait for you.
Called “POEMPORTRAITS,” the online collective artwork is a combination of poetry, design and Machine Learning (ML).
“A ‘POEMPORTRAIT’ is your self-portrait overlaid with a unique poem, created by AI. You can create your own and contribute to the evolving, collective poem,” Google said in a blog post on Thursday.
To create your poem portrait, donate a word of your choice and take a self portrait.
Each word you donate will be expanded into original lines of poetry by an algorithm that’s trained on millions of words of 19th century poetry.
“You’ll then receive a unique POEMPORTRAIT of your face, illuminated by your original lines of poetry. All of the lines of poetry are then combined to form an ever-evolving, collective poem,” said Google.
Es Devlin, a famous artist and stage designer, collaborated with Google Arts and Culture Lab, and coder Ross Goodwin for the project.
Ross trained an algorithm to learn to write poems by reading over 25 million words written by 19th century poets.
“It works a bit like predictive text: it doesn’t copy or rework existing phrases, but uses its training material to build a complex statistical model. As a result, the algorithm generates original phrases emulating the style of what it’s been trained on,” said Google. (IANS)
Google security researchers discovered several security flaws in a privacy software in Apple web browser Safari that could have helped third-party vendors track users’ browsing habits.
According to a report in the Financial Times which cited a soon-to-be published paper from Google’s ‘Project Zero’ team, the vulnerabilities were found in an anti-tracking feature known as ‘Intelligent Tracking Prevention’.
Once disclosed by Google researchers to Apple in August last year, the Cupertino-based iPhone maker immediately patched the flaws.
Apple launched the ‘Intelligent Tracking Prevention’ tool in 2017 to, in fact, protect Safari users from being tracked around the web by advertisers and other third-party cookies.
According to Google researchers, the vulnerabilities left personal data of Safari users exposed. They also found a flaw that allowed hackers to “create a persistent fingerprint that will follow the user around the web”.
Apple confirmed it patched the issues.
This is the third time Google researchers have found flaws in the Apple ecosystem.
In September, Apple slammed Google for creating a false impression about its iPhones being at hacking risk owing to security flaws that allegedly let several malicious websites break into its iOS operating system.
Researchers at ‘Project Zero’ team had discovered several hacked websites that allegedly used security flaws in iPhones to attack users who visited these websites — compromising their personal files, messages, and real-time location data.
In a statement, Apple said the so-called sophisticated attack was narrowly focused, not a broad-based exploit of iPhones “en masse” as described.
According to Google, the websites delivered their malware indiscriminately and were operational for years.
Apple said that it fixed the vulnerabilities in question — working extremely quickly to resolve the issue just 10 days after it learnt about it.
In July last year, the ‘Project Zero’ team found six critical flaws in Apple iMessage that can compromise the user’s phone without even interacting with them. These security vulnerabilities fell into the ‘interactionless’ category.
Two members of ‘Project Zero’, Google’s elite bug-hunting team, published details and demo proof-of-concept code for five of six ‘interactionless’ security bugs that impact the iOS operating system and can be exploited via the iMessage client. All the six security bugs were patched with the iPhone maker’s iOS 12.4 release. (IANS)