Google Gemini Director’s Disturbingly Racist Posts Discovered: This Explains Why Its AI Erased White People

In the week that Google’s new artificial intelligence platform launched, its seeming inability to produce images of white men regardless of the prompt made it an unmitigated disaster.

And with more information unearthed regarding the people responsible for Google Gemini’s insufferable wokeness, its deficiencies look much less like a bug and more like a feature.

As reported Thursday by the Babylon Bee’s news site, Not the Bee, the senior director of the project, Jack Krawczyk, addressed Gemini’s problems in a post on the X social media platform — before he locked his account.

“We are aware that Gemini is offering inaccuracies in some historical image generation depictions, and we are working to fix this immediately,” Krawczyk said.

Between the wording of his post and Google’s decision to suspend Gemini’s image-generating feature, it appears the tech giant wants the insanely diverse images to seem like a simple mistake.

After X users unearthed some of Krawczyk’s past posts, however, the inherent bias of the platform seems much less accidental.

Among those — shared by the @LeftismForU account — are gems such as, “White privilege is f***ing real,” “I don’t mind paying more taxes and investing in overcoming systemic racism,” and, “This is America, where racism is the #1 value our populace seeks to uphold above all.”

Keep reading

40,000 YEARS AGO, NEANDERTHALS DEVELOPED THIS STICKY TECHNOLOGY TO HELP THEM CRAFT STRONGER TOOLS

According to a new study, Neanderthals were using an ochre-based compound as a glue-like adhesive at the Mousterian type site in Le Moustier, France, around 40,000 years ago.

Led by Patrick Schmidt at Eberhard Karls University of Tübingen in Germany, and published in Science Advances, the study documents the earliest known use of a multi-component adhesive in Europe, indicating complex cognitive abilities and technological innovation among Neanderthals.

A STICKY SITUATION

The research team reexamined stone tools from Le Moustier, which had been untouched since the 1960s, and found traces of a mixture of ochre and bitumen on several artifacts, including scrapers, flakes, and blades. Ochre, a naturally occurring reddish earth pigment, and bitumen, a common component of asphalt, were combined to create a sticky glue-like mass for tool making. 

This mixture was found to be more than 50% ochre, which was surprising to the team because air-dried bitumen alone can serve as an adhesive, but loses its adhesive properties when mixed with such large proportions of ochre. So the team ran some tests.

Keep reading

NEW PHASE OF MATTER CREATED DURING EXPERIMENTS WITH EXOTIC PARTICLES IN QUANTUM PROCESSOR

A new phase of matter previously recognized only in theory has been created by researchers using a quantum processor, which demonstrates the control of an exotic form of particles called non-Abelian anyons.

Neither fermions nor bosons, these exotic anyons fall someplace in between and are believed only to be able to exist in two-dimensional systems. Controlling them allowed the creation of an entirely new phase of matter the researchers now call non-Abelian topological order.

THE WORLD OF NON-ABELIAN ANYONS

In our everyday world of three dimensions, just two types of particles exist: bosons and fermions. Bosons include light, as well as the subatomic particle known as the Higgs boson, whereas fermions comprise protons, neutrons, and electrons that constitute the matter throughout our universe.

Non-Abelian anyons are identified as quasiparticles, meaning that they are particle-like manifestations of excitation that persist for periods within a specific state of matter. They are of particular interest for their ability to store memory, which may have a variety of technological applications, particularly in quantum computing.

One of the reasons for this is because of the stability non-Abelian anyons possess when compared to qubits, which are currently used in quantum computing platforms. Unlike qubits, which can at times be less than reliable, non-Abelian anyons can store information as they move around one another without the influence of their environment, making them ideal targets for use in computational systems once they can be harnessed at larger scales.

In recent research, Ashvin Vishwanath, the George Vasmer Leverett Professor of Physics at Harvard University, used a quantum processor to test how non-Abelian anyons might be leveraged to perform quantum computation.

Keep reading

INVENTORS CREATE WORLD’S FIRST ‘WATER BATTERIES’ THAT WON’T CATCH FIRE OR EXPLODE

An international team of researchers and industry collaborators say they have invented the world’s first water batteries.

Unlike popular lithium-ion batteries, which power everything from mobile phones to electric cars but are made with dangerous and toxic electrolytes, these new batteries use only water. The result is an energy storage device that is less toxic, fully recyclable, and one that will never catch fire or explode.

Although the performance of water batteries is still short of lithium-ion batteries, their inventors say numerous advances and planned improvements should close that gap within the next five to ten years.

In a traditional battery, organic electrolytes allow electric current to flow between the positive and negative terminals. In lithium-ion batteries, a liquid solvent that dissolves lithium salt acts as the electrolyte. Unfortunately, these materials often don’t react well to high temperatures or pressures. In the worst cases, they can burn up electronic devices, set car engines on fire, or even explode.

Hoping to find a better alternative, an international team of research scientists led by RMIT University Distinguished Professor Tianyi Ma and selected industry partners figured out a way to replace this dangerous and often toxic electrolyte with plain old H2O, a.k.a. water. The result is an energy storage medium that works just like a lithium-ion battery but without all of the unwanted risks.

Keep reading

What The Rise Of AI Videos Means For The Future

OpenAI, a company backed by Microsoft, is one of the world’s leading artificial intelligence research organizations paving the way for AI advancement across multiple technology sectors.

Best known for the creation of ChatGPT, on February 15th the company debuted previews of its new prompt-based AI video generation software, Sora. A text-to-video tool which allows users to input a text prompt into the software which then uses generative AI to create a video based on the input.

While Sora isn’t the first of its kind as far as generative AI video tools are concerned, industry insiders have noted the crispness and detail as well as the ability to generate complex scenes and longer length videos than previous models, at present totaling up to sixty seconds in length, as a “significant leap” for the development of the technology.

On its website, OpenAI has stated “We’re teaching AI to understand and simulate the physical world in motion, with the goal of training models that help people solve problems that require real-world interaction.” While providing several brief examples of videos generated with the software and without any further modification.

The videos span a variety of scenes, including photorealistic clips of puppies playing in the snow, historic footage of California during the gold rush, a robot living in a cyberpunk world, woolly mammoths in the snow, an animation of a cute fluffy creature playing with fairies in an enchanted forest, and various cityscapes along with other animations.

Keep reading

Your fingerprints can be recreated from the sounds made when you swipe on a touchscreen — Chinese and US researchers show new side channel can reproduce fingerprints to enable attacks

An interesting new attack on biometric security has been outlined by a group of researchers from China and the US. PrintListener: Uncovering the Vulnerability of Fingerprint Authentication via the Finger Friction Sound [PDF] proposes a side-channel attack on the sophisticated Automatic Fingerprint Identification System (AFIS). The attack leverages the sound characteristics of a user’s finger swiping on a touchscreen to extract fingerprint pattern features. Following tests, the researchers assert that they can successfully attack “up to 27.9% of partial fingerprints and 9.3% of complete fingerprints within five attempts at the highest security FAR [False Acceptance Rate] setting of 0.01%.” This is claimed to be the first work that leverages swiping sounds to infer fingerprint information.

Biometric fingerprint security is widespread and widely trusted. If things continue as they are, it is thought that the fingerprint authentication market will be worth nearly $100 billion by 2032. However, organizations and people have become increasingly aware that attackers might want to steal their fingerprints, so some have started to be careful about keeping their fingerprints out of sight, and become sensitive to photos showing their hand details.

Without contact prints or finger detail photos, how can an attacker hope to get any fingerprint data to enhance MasterPrint and DeepMasterPrint dictionary attack results on user fingerprints? One answer is as follows: the PrintListener paper says that “finger-swiping friction sounds can be captured by attackers online with a high possibility.” The source of the finger-swiping sounds can be popular apps like Discord, Skype, WeChat, FaceTime, etc. Any chatty app where users carelessly perform swiping actions on the screen while the device mic is live. Hence the side-channel attack name – PrintListener.

Keep reading

Lawmakers Want Pause on Federal Funds for Predictive Policing

Should data scientists be in the business of fingering Americans for crimes they could commit, someday? Last month, a group of federal lawmakers asked the Department of Justice to stop funding such programs—at least until safeguards can be built in. It’s just the latest battle over a controversial field of law enforcement that seeks to peer into the future to fight crime.

“We write to urge you to halt all Department of Justice (DOJ) grants for predictive policing systems until the DOJ can ensure that grant recipients will not use such systems in ways that have a discriminatory impact,” reads a January letter to Attorney General Merrick Garland from U.S. Sen. Ron Wyden (D–Ore.) and Rep. Yvette Clarke (D–N.Y.), joined by Senators Jeff Merkley (D–Ore.), Alex Padilla, (D–Calif.), Peter Welch (D–Vt.), John Fetterman, (D–Penn.), and Ed Markey (D–Mass.). “Mounting evidence indicates that predictive policing technologies do not reduce crime. Instead, they worsen the unequal treatment of Americans of color by law enforcement.”

The letter emphasizes worries about racial discrimination, but it also raises concerns about accuracy and civil liberties that, since day one, have dogged schemes to address crimes that haven’t yet occurred.

Keep reading

PENTAGON’S BLUE PROGRAM AIMS TO FUEL UNDERWATER REMOTE SENSORS USING MICROSCOPIC MARINE ORGANISMS

The Defense Advanced Research Projects Agency (DARPA) has launched an ambitious effort to harness the power of microscopic marine organisms to power remotely deployed naval sensor platforms. 

The initiative, known as the “BioLogical Undersea Energy” or “BLUE” program, seeks to revolutionize the capabilities of ocean-deployed sensor technologies by developing self-refueling power supplies that run on dissolved organic matter abundantly found throughout the world’s oceans. 

“The BLUE program seeks to develop technologies to continuously provide electrical power that will expand the capabilities of remote, ocean-deployed sensor systems,” reads documents issued to prospective industry partners. “Such systems hold great potential for national security, understanding dynamics of marine environments, and monitoring marine climate change.”

According to a solicitation notice issued last week by DARPA, the BLUE program is driven by the recognition that current battery-powered sensor systems face limitations due to their finite energy capacity and frequent need for recharging or swapping of batteries.

Replacing batteries for some underwater sensors can present significant logistical hurdles and even risks to personnel and equipment. This is particularly evident in scenarios involving covert sensor systems that monitor strategic waters adjacent to a potential foreign adversary.

By exploring alternative energy sources derived from marine biomass, DARPA seeks to overcome these challenges and unlock new remote marine monitoring and surveillance possibilities.

Specifically, the BLUE program will target microscopic forms of marine biomass, including dissolved organic matter, phytoplankton, bacteria, and microscopic zooplankton, as potential electrical power sources. 

Unlike approaches utilizing macroscopic biomass, such as seaweed or kelp, BLUE will focus solely on harnessing energy from the abundant and diverse microorganisms in marine environments. 

“It is our hypothesis that the energy requirements of many ocean-deployed systems can be met by development of an onboard device that converts marine biomass into simple fuels and then converts those fuels into operational power,” Dr. Leonard Tender, BLUE program manager, said in a statement issued by DARPA. 

Using natural processes to convert organic waste into usable energy is hardly a novel idea.

Anaerobic digestion, a process where microorganisms break down biodegradable material without oxygen, is a significant source of renewable energy and biofuel production. The International Energy Agency reports that biofuels currently contribute to over 3.5% of the world’s transport energy, with projections indicating a 150% increase by 2030.

This is also not the first time the Pentagon has tried to recruit marine life to serve out U.S. national security interests. A previous DARPA effort, the “Persistent Aquatic Living Sensors” or “PALS” program, sought to use marine animal behavior as a way of monitoring strategic waters, including tracking adversarial subs. 

Nevertheless, DARPA’s aim to replicate biofuel production within a self-contained underwater system is unprecedented. While innovative, this ambitious endeavor could likely encounter some engineering hurdles, primarily due to the intricacies of operating in remote marine environments.

To meet program requirements, a power supply must sustain at least 0.1 kW average continuous power for over a year while remaining fully submerged. The device should also be compact enough to fit within the specified size and weight constraints of a form factor of less than 180 liters or 440 lbs. 

Crucially, the power supply should be expected to self-refuel on marine biomass, offering a persistent and sustainable energy solution for remote sensor systems in oceanic environments. 

As with most programs run by the Pentagon’s brain trust, DARPA does not elaborate on what times of sensor platforms a new microscopic marine power supply might fuel. However, solicitation documents repeatedly mention the need for systems to operate underwater and be capable of independently providing consistent power for “at least one year.” The microscopic marine organisms fueling the system must also be “sufficiently abundant” in locations identified through sources such as satellite imaging. 

Keep reading

SCIENTISTS CAN SEE AROUND OBSTACLES WITH THIS REALITY HACK THAT TURNS ORDINARY SURFACES INTO MIRRORS

Two scientists from the University of Southern Florida (USF) say they have developed a technique that allows them to see around obstacles.

According to the technique’s inventors, their algorithm, which was inspired by a car crash, can use a single photograph to compute “highly accurate, full-color three-dimensional reconstructions of areas behind obstacles.”

Although still in the development stages, the inventors say they could see several uses for their technique. Potential applications range from traffic safety and military operations to police forces navigating hostage situations.

“We live in a 3D world,” said the technique’s co-inventor, USF assistant professor of computer science John Murray-Bruce, “so obtaining a more complete 3D picture of a scenario can be critical in several situations and applications.”

THE ABILITY TO SEE AROUND OBSTACLES IS ALL IN THE SHADOWS

To create the algorithm, Murray-Bruce and his doctoral student, Robinson Czajkowski, focused on the information available in a digital photograph. Most importantly, they analyzed the data available from shadows cast by objects and structures that are obstructed from the camera’s point of view.

“These shadows are all around us,” Czajkowski said. “The fact we can’t see them with our naked eye doesn’t mean they’re not there.”

In fact, the pair of scientists say there is a ton of information in these shadows that the human eye simply cannot interpret. Their algorithm also incorporates data from “reflections” cast by objects obscured from the camera’s point of view.

Keep reading

NAVAL RESEARCH LABORATORY LEVERAGES EXOTIC PROPERTIES OF 2D WAVEGUIDES TO CAPTURE THE LIGHT OF DARK EXCITONS

A milestone discovery of waveguides based on two-dimensional materials with “exotic” properties has been achieved, according to researchers with the United States Naval Research Laboratory (NRL).

The breakthrough was achieved during studies involving hexagonal boron nitride, a two-dimensional material, by the U.S. Navy’s official corporate research lab while working with researchers from Kansas State University.

EXPERIMENTS IN THE REALM OF 2D MATERIALS

2D materials are a unique class of substances that result from the reduction of a material to its recognized monolayer limits through processes of mechanical peeling. This results in a unique class of ultra-thin 2D materials which have a variety of applications.

The most famous 2D material is graphene, which is formed from bonded carbon atoms to create a sheet that is just one atom in thickness. Such materials can be separated into layers by exploiting what is known as the van der Waals attraction, which describes an attracting force between atoms or molecules that is dependent on distance, rather than chemical bonding.

Following successful recent experiments with graphene, researchers have engaged in experimentation with other novel 2D materials, including hexagonal boron nitride (BN), the softest and most stable of BN’s forms which has already seen applications as a lubricant in machine components, as well as in some cosmetic products.

However, of greater interest to studies at the NRL are the ways these materials can be leveraged at the nanoscale level. This makes them particularly useful for developing applications where extremely thin optical and electrical components are necessary.

Hexagonal boron nitride is a particularly special case, given that past samples examined by NRL scientists revealed its potentially useful optical qualities. However, further studies soon revealed a surprise: that the compound possessed other hidden, and extremely promising capabilities.

Keep reading