UK government gains research access to big tech AI models

In a significant move, the UK government secures access to AI models from tech giants, aiming to enhance research and safety measures in the field of artif

UK Prime Minister Rishi Sunak recently unveiled a significant development in the country's AI oversight strategy during his speech at London Tech Week.

In an effort to enhance research and safety measures, notable technology companies such as Google DeepMind, OpenAI, and Anthropic have made a commendable commitment to granting the UK government "early or priority access" to their AI models.

Although the extent and nature of the data that will be shared with the UK government remain undisclosed at the moment, inquiries have been made to Google, OpenAI, and Anthropic to shed light on this matter.

UK Government seeks accountability, safety and transparency

This announcement follows the government's recent declaration of conducting an initial evaluation of AI model accountability, safety, transparency, and other ethical concerns.

Playing a pivotal role in this endeavour will be the country's Competition and Markets Authority. 

Moreover, the UK has demonstrated its dedication to investing an initial sum of £100m (approximately US$125.5m) to establish the Foundation Model Taskforce.

This taskforce will be responsible for developing "sovereign" AI that will drive economic growth in Britain while simultaneously minimising ethical and technical predicaments.

Industry leaders call for temporary suspension of AI development

The move by the UK government comes in response to mounting calls from industry leaders and experts for a temporary suspension of AI development.

These concerns stem from worries that developers are forging ahead without adequate consideration for safety measures.

Although generative AI models like OpenAI's GPT-4 and Anthropic's Claude have garnered praise for their vast potential, they have also sparked apprehension due to issues such as inaccuracies, misinformation, and potential abuses, including cheating.

Consequently, the UK's decision to gain direct access to these models represents a significant step towards mitigating these concerns and proactively addressing problematic models before they cause substantial harm.

It is important to note that this arrangement does not grant the UK government unfettered access to the entirety of these models and their underlying code.

Furthermore, it does not guarantee that every major issue will be promptly identified by the government. However, the access provided does hold the promise of yielding valuable insights.

At the very least, this initiative signifies a marked increase in transparency within the AI landscape, a crucial development given the current uncertainty surrounding the long-term impact of these systems.

Share

Featured Articles

The Dangers of AI Bias: Understanding the Business Risks

As Google seeks to fix AI bias issues within Gemini, concerns over AI and machine learning biases remain as developers consider combatting inaccuracies

Fujitsu to Combat 5G+ Network Complexities with AI

Fujitsu announces Virtuora IA, a collection of AI-powered network applications that used machine learning models to improve mobile network performance

MWC Barcelona 2024: The Power of AI in the Telco Industry

Mobile Magazine is live at MWC Barcelona 2024 this week! Discover some of the leading trends and businesses that are digitally transforming the industry

Upskilling Global Workers in AI with EY’s Beatriz Sanz Saiz

AI Strategy

Intuitive Machines: NASA's Odysseus bets on Private Company

Data & Analytics

Unveiling Gemma: Google Commits to Open-Model AI & LLMs

Machine Learning