
Kenyans, Ugandans gaining little from AI jobs despite high training
Despite being globally renowned suppliers of business process outsourcing (BPO) talent, Kenyans and Ugandans tapped to work in the AI industry often get low-skill and low-paying jobs, even though they are highly trained and skilled, discouraging the growth of talent in the sector.
ChatGPT – one of the world's most famous artificial intelligence (AI) chatbots – was trained mostly by Kenyans, and new research now shows may also have contributed to 'deskilling' the youth due to a mismatch of qualifications.
UN Trade and Development (Unctad) Technology and Innovation Report 2025 highlights the stark reality of AI workers in countries such as Kenya, Uganda and India, which are global leaders in BPO work.
A recent survey on micro task platforms and BPO companies showed that in Kenya and India specifically, 'highly educated workers, with graduate degrees or specialised educations in science, technology, engineering or mathematics, were often relegated to relatively low-skill tasks such as text and image annotation and content moderation.' 'Such significant wastes of human capital may be exacerbated in increasingly connected job markets, in which tasks are outsourced globally,' Unctad warns in the biennial technology report.'Data annotators in developing countries often experience difficult conditions, including up to 10 hours of work per day at wages of less than $2 per hour, engaged in repetitive tasks, and with limited opportunities for career advancement, for example, in Kenya and Uganda.'Essentially, such jobs are eroding Kenya's rich tech talent, and edging out its competitive advantage in the increasingly growing global tech industry, further widening the technology gap and consequently the income inequality gap.
Read: IMF: AI could worsen income inequalityDespite being highly trained and skilled in science, technology, engineering and mathematics, Kenyans struggle to find meaningful employment locally and internationally, forcing them to settle for the low-skill AI jobs offered by BPO companies.
Over the years, Kenya has developed more tech talent and is now one of the countries with the fastest growth in developers in Africa and globally.
Between 2022 and 2023, Kenya recorded the second fastest growth in the number of developers listed on coders platform Github at 41 percent, second only to Nigeria's 45 percent.
Last year, Kenya posted the fastest growth in Github developers at 33 percent, pushing the total number of programmers in the country to over 393,000, the fifth highest in Africa after Nigeria, Egypt, South Africa, and Morocco.
Also read: Jobs that will survive AI, and those that won'tThe country is also one of the top four tech start-up giants on the continent, a testament to leading in tech talent and innovation, but this has also been changing over time, with such start-ups starved of cash from investors forcing many of them to scale back operations or fold completely.
Data from the African Venture Capital Association shows that last year, total funding raised by Kenyan start-ups declined by 33 percent to $318 million from $473 million in 2023.
As a result, at least five start-ups closed down after failing to raise follow-up funding, leaving hundreds of talented Kenyans jobless amid tightening economic conditions and falling income.
This came hot on the heels of similar scale-backs and shutdowns in 2023, which also left many in the country without jobs. To protect workers in the tech industry, Unctad proposes key policy changes that need urgent action by State organs.'Translating technological progress into shared prosperity requires labour-friendly policies in three stages: investments in education and skills, in pre-production; labour protection and worker empowerment, in production; and progressive taxation, in post-production,' the UN agency said in the report.
© Copyright 2022 Nation Media Group. All Rights Reserved. Provided by SyndiGate Media Inc. (Syndigate.info).
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Zawya
a day ago
- Zawya
OpenAI to continue working with Scale AI after Meta deal
OpenAI plans to continue working with Scale AI after rival Meta on Friday agreed to take a 49% stake in the artificial intelligence startup for $14.8 billion, OpenAI's CFO Sarah Friar told the VivaTech conference in Paris. Scale AI provides vast amounts of labelled or curated training data, which is crucial for developing sophisticated tools such as OpenAI's ChatGPT. "We don't want to ice the ecosystem because acquisitions are going to happen," she said. "And if we ice each other out, I think we're actually going to slow the pace of innovation." (Reporting by Supantha Mukherjee in Paris, Editing by Louise Heavens)


Arabian Post
a day ago
- Arabian Post
macOS Embraces Linux Containers with Native Support
Apple has unveiled a breakthrough open‑source framework, Containerization, during its WWDC 2025 keynote, enabling developers to create, run and manage Linux containers directly on macOS. The Container CLI, a companion command‑line tool, operates each container as a lightweight virtual machine, bypassing the need for third‑party platforms like Docker. This marks a strategic shift in Apple's support for cross‑platform workflows, particularly for developers working on server‑side and cloud‑native applications. Containers have become central to modern software engineering by packaging applications with all dependencies in a consistent, portable environment. Up to now, Mac users have typically relied on resource‑heavy, shared VMs to run Linux containers, often encountering sluggish performance and battery drain on Apple Silicon machines. Apple's solution leverages its own Virtualization framework and Apple Silicon optimisations to spin up sub‑second containers, each within its own minimal‑footprint VM. Isolation and security are core pillars of Containerization. Each container receives a dedicated IP address, entirely separate CPU and memory allocations, and performs directory sharing only when explicitly requested. The container's root filesystem omits core utilities, libc, and dynamic libraries by default — a deliberate measure to reduce the attack surface. The init process, vminitd, is written in Swift and acts as the VM's first process, handling IP assignment, filesystem mounting and process supervision. ADVERTISEMENT Performance gains are significant. By optimising the Linux kernel and exposing container filesystems as EXT4 block devices, Apple has achieved rapid cold‑boot speeds while maintaining low I/O overhead. Benchmarks suggest these containers outperform Docker Desktop in terms of startup time, memory footprint and CPU use on Apple Silicon systems. Technical details from GitHub show support for OCI‑compliant images, enabling compatibility with existing registries and Kubernetes systems. The container CLI mirrors familiar commands—pulling and running Alpine Linux images is as simple as typing container image pull alpine:latest. The project repo, licensed under Apache‑2.0, is written entirely in Swift, optimised for Apple Silicon and designed for community contribution. Apple's launch places Containerization among strong open‑source contenders such as Podman, containerd, Buildah and Rancher Desktop. Yet the per‑container VM model marks a departure from standard shared‑kernel container runtimes, offering enhanced isolation at the cost of slightly increased base resource use. Notably, critics and users on Reddit and in industry commentary have voiced curiosity about whether this approach outclasses lightweight VM tools like Orbstack or Lima. A potentially limiting factor is network isolation on macOS 15. Full network capabilities, including container‑to‑container traffic, are only available on the upcoming macOS 26, currently in beta and expected later in the year. Users on the earlier Sequoia release may experience restricted container networking or compatibility issues. Additionally, a Rosetta 2 bug affecting x86_64 processes in Linux VMs may impede workflows involving amd64 containers—a challenge both Apple and downstream projects like Podman are working to resolve. The timing of Containerization aligns with a broader developer toolkit refresh. Alongside this framework, Apple introduced Swift 6.2, Xcode 26 featuring LLM integration, and Game Porting Toolkit 3. This suite reflects a strategic push to consolidate development workflows across desktop, mobile, AI and cloud environments within the Apple ecosystem. Early adopters with Apple Silicon and access to macOS 26 beta are already testing the CLI and framework. Feedback is mixed: some praise the speed and security enhancements, while others caution that lack of full networking and Rosetta issues may restrict use in complex container orchestration setups. Apple's decision to open‑source Containerization is notable. It invites cross‑platform contributions, and standards compatibility via Swift and OCI means downstream projects could integrate with or build on the framework. If momentum grows, it could prompt a shift away from third‑party container tools on macOS, benefiting the entire developer ecosystem.


Khaleej Times
2 days ago
- Khaleej Times
UAE: ChatGPT outage; OpenAI identifies root cause, working towards full recovery
ChatGPT has identified the root cause for the issue causing elevated errors and latency across the listed services, OpenAI said, adding that it is working on implementing a mitigation. There is continued recovery across API and ChatGPT, and engineers have applied the mitigation and are monitoring the recovery. However, they are still monitoring the fix and are working towards full recovery. Users in the UAE reported issued while using ChatGPT as thousands across the globe were hit with an OpenAI outage on Tuesday, as per information on data-tracking website Downdetector. From the reported issues globally, around 93% of users faced problems while using ChatGPT; seven per cent had issues while using the application and a percentage of users were struggling while trying to log into the site. . The technical issues peaked at 1.30pm, when most users reported issues. At 2.45pm, reports had gone down but were still high around the world. In the UAE, reports peaked at 1.40pm and have been ongoing until now. Around 60% of Emirates' users faced issues while trying to use the website; 38 per cent of individuals had issues while logging in and two per cent of users had issues with ChatGPT. Speaking to Khaleej Times, Sarah K., a Dubai resident, said she was trying to compile her university notes but was unable to do so. "I have an assignment due at 5pm and just wanted to compile the notes together, but the page is unresponsive. It feel weird to not get a response from ChatGPT," said the 19-year-old. Earlier this year, ChatGPT removed the sign up option for users, making it simpler for individuals to log in. OpenAI initially launched the search engine to paid subscribers. It soon made it accessible to all users by rolling it out in the same year. Users still required an account to access it, as without one it would only allow for an allocated number of tries for a certain duration. This move is likely to put pressure on search engines like Google and Bing as the platform will now be easily accessible to all.