Telegram (AI) YouTube Facebook X
Ру
An algorithm developed chemical weapons, a neural network created a coronavirus vaccine, and other AI news

An algorithm developed chemical weapons, a neural network created a coronavirus vaccine, and other AI news

We aim to inform readers not only about events in the bitcoin industry, but also about developments in adjacent technological fields — cybersecurity and the world of artificial intelligence (AI).

ForkLog AI has gathered the most important AI news from the past week.

  • A startup developed an algorithm to track companies that stopped operating in Russia.
  • A bipartisan bill was introduced in the United States, expanding transparency requirements for data collection.
  • Israel and Singapore signed a cooperation agreement in the field of artificial intelligence.
  • The development of autonomous freight transport will displace 500,000 American jobs.
  • Alphabet spun off the Sandbox Group quantum computing developers into a separate company.
  • AI, in six hours, proposed 40,000 new potential units of chemical weapons.
  • Pfizer employed AI to develop a coronavirus vaccine and medicines.
  • The most notable AI deals from the past week.

A startup developed an algorithm to track companies that stopped operating in Russia

The Stanford startup Diffbot developed an AI algorithm to track companies that ceased operating in Russia due to the war in Ukraine. As of today, about 650 organisations are listed in the sanctions tracker.

Using NLP, the system scans the internet and structures the data into a publicly accessible Knowledge Graph. The algorithm also assigns each source a trust rating and probability scores for each stated fact, cross-checking and consolidating information from other outlets.

CEO Mike Tang said Diffbot monitors sanctions in real time by reviewing data that appears in at least three independent credible sources.

Ukraine will use facial recognition to locate relatives of dead Russian soldiers

The authorities of Ukraine will deploy facial recognition to identify Russian military personnel who have invaded the country.

According to Digital Transformation Minister Mykhailo Fedorov, the algorithm will match photographs of the deceased soldiers with images on social networks. In the future, the agency plans to establish contact with their relatives.

In the US, data-collection transparency requirements for the government will be expanded

American congressmen introduced a bipartisan bill expanding transparency requirements for data collection.

Under current norms, the government is required to notify citizens about wiretaps and the review of bank account records. Lawmakers proposed expanding this list to include email correspondence, text messages, location data, and cloud data.

“When the government obtains someone’s emails or other digital information, people have the right to know about it,” said Senator Ron Wyden.

The bill would also compel authorities to publish information about every surveillance order. Law enforcement agencies would be required to inform the court about a search of a home or device not belonging to the person under investigation, as well as the improper disclosure of personal data by tech companies.

Israel and Singapore signed a cooperation agreement in AI

The governments of Israel and Singapore signed a cooperation agreement in AI. The authorities of both countries agreed to intensify collaboration in AI technologies.

According to Singapore’s Minister of Foreign Affairs Vivian Balakrishnan, the agreements will support the development and deployment of AI. He also announced the opening of an embassy in Tel Aviv to “strengthen the innovative partnership between the two countries.”

Study: autonomous freight transport will displace 500,000 American workers

Researchers from the University of Michigan said the development of autonomous freight transport will displace 500,000 jobs in the US.

According to the researchers, this represents about 90% of long-haul truck drivers in the United States.

They noted that there are a number of barriers to automation of freight transport, including the inability of algorithms to navigate in bad weather, and the lack of a legal framework and appropriate infrastructure.

Boeing creates an AI combat drone

The Boeing company unveiled a Ghost Bat, an AI-powered combat drone.

Drone Ghost Bat. Data: Boeing.

The drones are produced in Australia and designed to fly alongside manned fighters. Ghost Bat can suppress radar signals, monitor terrain, or engage targets.

“As air warfare becomes more complex and sophisticated, drones are becoming a vital asset for the armed forces,” said Australian Defence Minister Peter Dutton.

Alphabet spun off its experimental quantum division into a separate company

Alphabet spun off the Sandbox Group’s quantum computing developers into a separate company, Sandbox AQ. It will blend quantum computing and AI with a focus on “post-quantum cryptography,” data security, and green energy.

The business is built around a set of APIs called Floq, allowing developers to use tensor processing units (TPU) to model workloads for quantum computing. Its source code is available on GitHub.

The new company is led by former Sandbox Group head Jack Hidary.

Neural network developed 40,000 types of chemical weapons in six hours

The AI proposed 40,000 new possible units of chemical weapons in just six hours. Researchers switched the drug-search algorithm to “bad actor” mode to demonstrate how easily it could be abused.

Instead of filtering out toxic compounds, the scientists asked the AI to search for them. The algorithm generated tens of thousands of new substances, some resembling VX — the most potent nerve agents ever developed.

According to lead author Fabio Urbina, any data scientist with expertise in chemistry could repeat the experiment.

“I don’t want to panic by saying there will be AI-powered chemical warfare. I don’t think this will happen soon. However, it is starting to become possible.”

AI helped Pfizer develop a coronavirus vaccine and medicines

The pharmaceutical company Pfizer employed AI and supercomputers at every stage of developing and delivering the COVID-19 vaccine. AI was also used in creating pills for the coronavirus.

Scientists were able to sequence and assemble the structure of the COVID-19 spike protein at the early stage of the pandemic. According to the researchers, the development of the mRNA vaccine Comirnaty and its clinical trials took just four months.

Pfizer used AI to forecast yield during production before tens of thousands of volunteers from six countries were recruited for testing. AI systems were used to analyze any discrepancies in participants’ symptoms.

Today the company uses machine learning to forecast product temperatures and to maintain more than 3,000 freezers storing vaccines. In addition, Pfizer uses the Internet of Things and sensors to track shipments and monitor temperatures with nearly 100% accuracy, scientists said.

Most important AI deals of the past week

From March 20 to 26, 2022, AI startups raised over $1.065 billion. Here are the most interesting deals.

  • The smart robotic window cleaners makers Skyline Robotics raised $6.5 million.
  • Datagen raised $50 million to develop synthetic data for computer vision systems.
  • MixMode raised $45 million to expand its self-learning platform for proactive protection against previously unknown cyberattacks.
  • Terra Drone raised $70 million to scale drone software.
  • Ramp raised $750 million.
  • Gensyn raised $6.5 million to launch a decentralized compute network for AI model training.
  • Finally raised $95 million to lend to small and medium businesses.
  • Apollo Agriculture raised $40 million to expand its customer base.
  • Whale Dynamic raised $2.5 million to explore robotic courier applications.

Also on ForkLog:

Subscribe to ForkLog News on Telegram: ForkLog AI — all AI news!

Подписывайтесь на ForkLog в социальных сетях

Telegram (основной канал) Facebook X
Нашли ошибку в тексте? Выделите ее и нажмите CTRL+ENTER

Рассылки ForkLog: держите руку на пульсе биткоин-индустрии!

We use cookies to improve the quality of our service.

By using this website, you agree to the Privacy policy.

OK