OpenAI and Malta’s Nationwide ChatGPT Plus Partnership

OpenAI announced on May 16, 2026, a groundbreaking partnership with the Government of Malta to provide ChatGPT Plus to all citizens, marking the world’s first nationwide initiative of its kind. This “AI for All initiative” will enable all Maltese citizens to access ChatGPT Plus for free.

According to George Osborne, Head of OpenAI for Countries, “Malta is leading Europe and the world in making AI available to all its citizens. Intelligence is becoming a national utility, and all governments have a crucial role to play in ensuring their citizens have access to AI and the skills to use it to its full potential.”

The program will be managed by the Malta Digital Innovation Authority, with the first phase launching in May 2026. Participants will have access to a free one-year subscription to ChatGPT Plus after completing an AI literacy course developed by the University of Malta.

(Source: OpenAI and Malta partner to bring ChatGPT Plus to all citizens)

AI Literacy Education and Practical Implementation

Malta’s AI for All initiative adopts an approach that prioritizes education over mere technology provision. The course developed by the University of Malta covers basic AI concepts, what AI can and cannot do, and responsible usage in households and workplaces.

As explained by Silvio Schembri, Malta’s Minister for the Economy, Enterprise, and Strategic Projects, “Through this AI for everyone course, we are ensuring that all citizens, regardless of background, have the confidence and skills needed to succeed in the digital world.” By combining education with free access to cutting-edge digital tools, the program transforms unfamiliar concepts into practical support for families, students, and workers.

The program is expected to expand in scale as more Maltese residents and citizens living abroad complete the course.

(Source: OpenAI and Malta partner to bring ChatGPT Plus to all citizens)

Practical Methods for Sales Teams Using Codex

OpenAI has released specific workflows for sales teams using Codex. Codex consolidates sales context scattered across CRM fields, call records, email threads, Slack discussions, decks, customer documents, and account signals to create a working draft of practical deliverables.

Key use cases include pipeline prioritization from unengaged accounts, meeting preparation and follow-up, forecast review and commit risk monitoring, strategic account plan updates, and diagnosis of stalled deals. Sales representatives and managers are responsible for relationship strategy and judgment, while Codex presents a working draft to the team earlier.

Codex integrates with plugins like Gmail, Slack, Gong, Google Drive, Spreadsheets, and Documents to generate specific first drafts of preparation briefs, follow-up notes, forecast risk notes, account plans, and escalation plans based on account history, customer conversations, deal signals, open risks, and review expectations.

(Source: How sales teams use Codex)

Performance Improvement through Asynchronous Continuous Batching

Hugging Face has announced an asynchronous technology that enables parallel execution of CPU and GPU in continuous batching processing. Traditional synchronous batching processing resulted in idle time as the GPU waited for the CPU to prepare the next batch and vice versa.

With the H200 costing around $5 per hour on Inference Endpoints, maximizing GPU utilization is crucial. In loops operating at hundreds of steps per second, these idle gaps can accumulate to occupy about a quarter of the total execution time.

Asynchronous batching uses CUDA streams and CUDA events to separate CPU batch preparation and GPU batch computation, allowing both to run in parallel. This enables the GPU to perform calculations continuously at 100% of the time. The implementation addresses technical challenges such as distinguishing between default and non-default streams, avoiding race conditions, and handling carryover processing.

(Source: Unlocking asynchronicity in continuous batching)

Summary

  • By referencing Malta’s AI for All initiative, organizations can design an introduction strategy that combines AI literacy education programs and access to practical tools across the entire organization.
  • Utilizing Codex’s sales team workflow, businesses can build a system that automatically generates pipeline analysis, meeting preparation materials, and forecast reviews from CRM data and customer context, enhancing operational efficiency.
  • Implementing Hugging Face’s asynchronous continuous batching technology can maximize GPU utilization in LLM inference servers, improving the cost-performance of high-cost GPUs like the H200.
  • By following OpenAI API best practices for production environments, organizations can securely operate scalable AI applications by properly configuring organization settings, billing limits management, and API key security.