Saturday 13 July 2024
EN    SK    RU

To become comparable to human intelligence, full AI must be able to transfer lessons learned from one domain to another, have common sense, cooperate with other machines or humans, and have self-awareness. Experts have different opinions about when humanity will create full AI: predictions range from 2030 to 2050. Many companies are developing artificial intelligence similar to human intelligence, but Open AI, Google Brain, DeepMind, and Facebook A.I. Research are leading the way.

How to recognize full AI

* Fool me if you can. The most famous route is the Turing test. Under the terms of the definitive test, the questioner communicates with two interlocutors, a machine and a human, if it cannot correctly determine which device has passed the test.

So far, no artificial intelligence has been able to achieve complete success.

* A cup of coffee in a stranger's house. Apple co-founder Steve Wozniak came up with his own full AI test: the machine passes if it can go into any room and find how to make itself a cup of coffee.

* Ivy League AI. This test requires the AI to go to a university and graduate using the same materials as other students.

* Taking people's jobs. This test requires the AI to do any primary job that humans can do.

Whether humanity should create a full AI

It's an open question. Physicist Stephen Hawking, Tesla founder Elon Musk, and other scientists and experts in the field have talked a lot about the risks of creating autonomous artificial intelligence.

So before full AI can emerge, humanity must decide how to regulate it.

More than 30 countries already have relevant regulations, the most famous being the U.S. and the EU. In September 2021, China also named its ethical guidelines for artificial intelligence.  

In 2016, Amazon, Facebook, Google, DeepMind, Microsoft, and IBM founded the Partnership on AI, a coalition of 95 companies that responsibly use artificial intelligence. In 2020, the team created the AI Incident Database.

Current state of the AI market

The global AI market was worth $62.35 billion in 2020 and is projected to reach $360.36 billion in 2028. McKinsey Global Institute expects AI's contribution to the global economy could reach $13 trillion by 2030.

In the first two quarters of 2021 alone, global investment in AI startups reached $38 billion, up from $670 million in 2011.

IBM, the largest owner of AI technology patents, owned 5,538 patent families in AI and machine learning in 2020. Its closest competitors, Samsung and Microsoft, had 5,500 and 5,052 active patent families.

In 2020, 50% of the companies McKinsey surveyed said they had implemented AI technology in at least one business area. Of those, two-thirds reported profit growth, and nearly half reported cost reductions. Accenture assures that AI technologies can increase business productivity by 40%.

Key players in the AI market 

In 2019, the Center for Data Innovation analyzed the U.S., Chinese and EU AI capacity in six categories: talent, research, development, hardware, implementation, and data. The U.S. led the way in four categories, scoring 44.2 out of a possible 100, while China led in two (32.3 points).

ITIF re-evaluated the same metrics in 2021. The U.S. scored 44.6, China scored 32.0, and the EU scored 23.3.

China is actively catching up with the U.S. in the development of AI: while in 2017, the gap between countries in the sector was 11 points, in 2019, it has already narrowed to six. Even the Pentagon's former director of software, Nicolas Chailan, publicly complained that China would win the race for leadership in the sector.

It is precisely what China's leadership aims for: in the fall of 2021, the state announced plans to become a leader in AI by 2030. China has enough resources to achieve this goal:

One of the strongest economies in the world.

Vast amounts of data (China's population is 1.4 billion people).

Breakthrough technologies.

The ability to focus these resources on the goal set.

China is not yet in the top 10 in terms of the readiness of states to implement AI in any of the authoritative (Western) rankings. It only loses points for "imperfect regulation," so soon China will move up from 19-20 positions in these rankings.

Applications of AI

AI is so deeply embedded in humanity's daily life that sometimes we don't even notice it. CB Insights compiles an annual list of the best AI companies. This study beautifully demonstrates that AI is transforming almost every aspect of civilization.

Machine learning, NLP, deep learning, and other AI technologies have made their way into people's lives on every level, from voice assistants to algorithm-driven stem cell synthesis. And this is far from the limit of how they can change the development of human civilization.

Li Kaifu, a Taiwanese investor and one of the leading AI experts, identified six trends that will shape the development of the artificial intelligence industry over the next 20 years:

◼ AI is driving innovation in innovative and automated businesses.

◼ Accelerating the emergence of the transportation of the future through intelligent cities and the Internet of Things.

◼ AI and digitalization will change the state of healthcare and human well-being.

◼ Robots bring the era of automation and intelligent manufacturing closer.

◼ Energy and agribusiness will depend on industrial production rather than natural resources.

◼ The cost of materials, energy, and production will fall considerably, and an "era of prosperity" will ensue.


FacebookMySpace TwitterDiggDeliciousStumbleuponGoogle BookmarksRedditNewsvineLinkedinRSS FeedPinterest
Pin It