At the tail end of a 3-day marathon negotiation, the European parliament and council leadership reached a ‘provisional,’ milestone agreement on harmonized rules governing the use of artificial intelligence.

This landmark agreement comes after years of sensitive negotiations since April 2021—when the framework was first brought forward.

Here’s a timeline of events since then, shared by the European Commission:

June 2021:

  • European Commission initiates a public consultation on Civil liability, focusing on adapting liability rules to the digital age and artificial intelligence.
  • European Commission proposes a Regulation on Product Safety.

November 2021:

  • Council of the EU, during the SI Presidency, produces a compromise text on the AI Act.
  • High-Level Conference on AI: From Ambition to Action is held, marking the 3rd European AI Alliance Assembly.
  • European Economic and Social Committee provides an opinion on the AI Act.

December 2021:

  • Committee of the Regions issues an opinion on the AI Act.
  • European Central Bank offers an opinion on the AI Act.

June 2022:

  • Spain launches the first AI regulatory sandbox, accelerating the implementation of the AI Regulation.

September 2022:

  • European Commission introduces a Proposal for an AI liability directive.

December 2022:

  • The Council adopts a general approach on the AI Act

June 2023:

  • European parliament defines its negotiating position on the AI Act

Carme Artigas, the Spanish Secretary of State for Digitalisation and Artificial Intelligence, contends that the provisional AI Act does a commendable job of striking a balance in the development of AI across Europe. This is achieved while upholding the rights of its citizens and ensuring their safety.

In a press release from the European Council, she states, “This is a historic achievement and a significant milestone for the future! Today’s agreement effectively addresses a global challenge in a rapidly evolving technological environment, focusing on a key area for the future of our societies and economies. In this endeavor, we managed to maintain an extremely delicate balance: fostering innovation and the adoption of artificial intelligence across Europe while fully respecting the fundamental rights of our citizens.”

The AI Act, which focuses on eight key elements, including the “Classification of AI systems as high-risk and prohibited AI practices,” will not apply to areas outside  EU law and AI systems being used exclusively for military or defense purposes.

Non-compliance with the AI Act, once it’s fully established, attracts fins anywhere between £7.5 million, or 1.5% of global turnover, and £35 million, or 7% of global turnover.

However, the EU asserts that the forthcoming laws will be flexible and can evolve. The framework includes clauses enabling the commission to adapt it as needed.

As the first global power to establish an AI governance framework, the EU stands as a benchmark for other governments that are likely to establish AI laws of their own.

It’s Gemini Season!—No, This Is Not About Astrology

In a year dominated by AI-centric innovation, every other news cycle introduces us to a new iteration or, at least, a fresh use case of the technology. Since the boom first sounded, we’ve seen AI paint, write the Great Gatsby set in a zombie apocalypse, code, work as I__nterns__, and now, understand real-time content like videos and human speech.

Gemini, Google’s latest AI model, is the company’s move to reaffirm its position as a leader in AI research. Since losing the spotlight to Microsoft-backed OpenAI, following ChatGPT’s launch last year, the Alphabet-owned company has tried to build and buy, its way back to the AI throne.

However, CEO Sundar Pichai believes that, with their newest large language model, the company finally has things figured out.

Wallstreet appeared to agree, as Alphabeth added $80 billion to its value last Thursday, following the public launch of Gemini AI.

The AI system–said to be faster than OpenAI’s latest model–comes in three versions, each designed with different processing power needs. Gemini Nano, a lighter version, is intended for native and offline use on Android devices. The other two public versions are Gemini Pro and Gemini Ultra.

Despite setbacks, especially after Bard shared inaccurate information that cost the company $100 billion in February, Google still has a soft spot for its favorite ex. For one, Bard is now powered by Gemini Pro, and Google has plans to roll out more advanced Gemini models through its flagship conversation chatbot next year.

In this whirlwind of rapid developments, the AI race remains an open contest. Yet, it’s evident that OpenAI and Google are enjoying a robust lead. *
Who dares challenge them?*

Google held a top-5 spot in HackerNoon’s weekly Tech Company Rankings at #4


In Other News…

  • You Should be Skeptical about QR Codes, FTC Warns – The Verge
  • Epic Games Wins Antitrust Lawsuit against Google – Tech Crunch
  • SpaceX postpones planned launch of US military’s secretive X-37B spaceplane – Reuters
  • New Mexico lawsuit accuses Meta of creating ‘breeding ground’ for child predators –  CNN
  • Teslas will be able to automatically call 911 if you get in a Crash –  The Verge

This brings us to the end of this week’s Tech Company News Brief.

Don’t forget to share this newsletter with your family, friends, and everyone in between!

See you next week!
Asher Umerie, Editor, World News & Scifi @ HackerNoon.

This article was originally published by Asher Umerie on Hackernoon.