With the European Union’s Artificial Intelligence Act (AI Act) enforcement window rapidly approaching, AI providers—from global tech giants to agile startups—are sounding the alarm. Scheduled to take effect on August 2, 2025, the Act introduces sweeping obligations for general-purpose AI (GPAI) models, including transparency, training data disclosure, copyright compliance, and risk management. Yet persistent uncertainties around implementation have triggered widespread concern over legal gray areas that threaten innovation and breed compliance risk.

The AI Act Implementation Timeline

• August 1, 2024: AI Act formally entered into force.
• February 2, 2025: First wave of rules activated—prohibitions on certain high‑risk AI practices and AI literacy obligations.
• May 2, 2025: Initial target date for the EU’s voluntary Code of Practice for GPAI, now widely delayed.
• August 2, 2025: GPAI‑related obligations kick in, including documentation, risk assessment, and transparency rules.
• August 2, 2026: Enforcement for high‑risk AI systems begins, subject to fines.
• August 2, 2027: Final compliance deadline for AI systems already in operation before 2025.

Essential supporting elements—like technical standards and the formal Code—remain unfinished or delayed, creating a murky compliance terrain.

What Is the Code of Practice — and Why It’s Delayed

A voluntary Code of Practice for GPAI models is intended as a practical blueprint for navigating the Act’s rules on transparency, copyright, safety, and security. While not legally binding, its adoption signals a pathway toward regulatory goodwill.

Originally slated for release in May 2025, the Code is now postponed—likely until late 2025 or even August—due to negotiations over content, stakeholder input, and legal alignment. Without the Code, providers must interpret vague legal language under time pressure.

Companies that are uncertain how to apply the rules—particularly surrounding copyrighted datasets and transparency obligations—face exposure to fines up to 7% of global annual revenue, or €35 million, depending on the infraction.

Legal Ambiguities Fuel Provider Frustration

Copyright vs. Transparency

AI firms remain unclear about how to attribute or document copyrighted materials in their training datasets. Explicit expectations regarding metadata disclosure and provenance are yet to be formally codified in technical guidance.

Scope of Obligations

Startups and cloud providers alike are grappling over who qualifies as a “provider” or “deploy­er” under the Act, and what constitutes a “general‑purpose” vs. a “high‑risk” model. That distinction deeply affects their regulatory responsibilities.

Technical Standards Gap

Essential harmonised standards that underpin compliance mechanisms—risk frameworks, testing procedures, security-by‑design protocols—are not yet available. Early drafts exist, but validation and endorsement remain pending.

Industry Pushback Grows

European Corporates Weigh In

Over 45 major European firms—including Airbus, BNP Paribas, Philips, SAP, and Spotify—have urged the EU to freeze or delay enforcement for up to two years, citing unclear guidance and escalating compliance costs.

Tech Giants’ Split Reactions

• Meta has refused to sign the Code of Practice altogether, with Chief Global Affairs Officer Joel Kaplan calling it an “overreach” laden with legal uncertainty.
• Google has committed to signing the full Code, though it publicly noted concern about the burdens imposed by copyright and data transparency rules.
• OpenAI, Mistral, and Microsoft have indicated support for the Code.
• xAI (Elon Musk’s company) plans to sign only the safety & security chapter, rejecting sections on copyright and transparency.

EU’s Official Response: Firm but Flexible

The European Commission has taken a firm stance: “there is no pause, no clock-stop, no grace period.” Prospective enforcement dates are legally binding.

Still, at a digital ministers’ meeting in Luxembourg, Executive Vice President Henna Virkkunen acknowledged the possibility of postponing implementation of select provisions if supporting standards are not finalised in time.

The Commission has promised simplifications—particularly for SMEs—later this year and plans to launch an AI Act Service Desk in summer 2025 to assist organisations with compliance planning.

What Providers Are Doing Now

Compliance Bootcamps

Firms are ramping up internal documentation: logging data sources, annotating models, building explainability layers, and aligning with GDPR. Risk registers and compliance matrices have become central to project management.

Legal Scenario Planning

Legal teams are drafting policies around copyright exceptions (e.g. fair use), preparing to justify missing provenance, and planning for disclosing disproportionate burden cases, particularly for legacy models already trained.

Stakeholder Dialogue

Industry bodies—including CCIA Europe, AI Champions, and VCs—are engaging in ongoing dialogue with Brussels on softening unclear obligations and aligning expectations.

Why EU AI Act Matters

• Global Tech Leaders Are Watching: The EU AI Act is the first comprehensive AI regulation globally. Its success—or failure—will influence policymaking in the U.S., UK, China, and beyond.
• European Innovation at Risk: Critics warn that poorly defined obligations may stifle EU startups, create regulatory arbitrage, and favour U.S./Chinese AI platforms.
• Risk of Scaled Enforcement Gaps: Without finalised standards, providers may over-interpret the law or under-prepare, and could face heavy penalties for non-compliance.
• Transparency vs. Trade Secrets: Firms must balance disclosure with intellectual property concerns. The tension between providing accountability and protecting innovation is front and centre—hence the label “legal grey areas.”

What’s Next

Code of Practice Finalisation

Whether the EU publishes the Code in late 2025 or just before the August 2 deadline remains key. Providers should monitor the European AI Office closely.

Industry‑EU Mediation

Expect further engagement between regulators and industry associations to refine implementation guidance and clarify definitions, especially around GPAI vs. high‑risk systems.

Simplification Proposals

By the end of 2025, the Commission intends to propose simplifications to ease the regulatory burden for smaller firms. These could take the form of adjusted reporting thresholds, tiered obligations, or phased reporting timelines.

Support Structures

Formal resources—such as the AI Act Service Desk, FAQs, and guidance documents—will become available. Companies must integrate these with existing compliance frameworks like ISO, GDPR and NIST.

Conclusion

The countdown to August 2, 2025 is more than a calendar reminder—it’s a pressure cooker for tech providers balancing innovation and regulatory compliance. As critical gaps in guidance and legal scope persist, the legal grey areas around training data, transparency, and high-risk classification threaten not only business continuity but Europe’s competitive edge in AI.

Providers must act now—documenting responsibly, engaging proactively, and planning legally robust strategies. Uncertainty may be unavoidable, but preparedness is key. The EU AI Act is unprecedented, and how businesses navigate its ambiguities today may shape the future of AI governance globally.

Join our WhatsApp community

Join Our Social Media Channels:

WhatsApp: NaijaEyes

Facebook: NaijaEyes

Twitter: NaijaEyes

Instagram: NaijaEyes

TikTok: NaijaEyes

READ THE LATEST TECH  NEWS

Crypto Market Sees Major Turmoil as Bitcoin, Ethereum Prices Plummet

The crypto world is in a state of high alert after a brutal week that saw major cryptocurrencies like Bitcoin and Ethereum suffer significant losses. This downturn has wiped out over $708 million from the market, with more than 160,000 traders getting “liquidated”—a term for when over-leveraged investors are forced to sell their assets due to the sharp price drops.

Here’s a simple breakdown of what happened:

Bitcoin’s Plunge: After almost reaching an impressive $120,000 just a few days ago, Bitcoin’s value has fallen to around $114,000, a drop of about 5.6%.

Ethereum’s Big Slide: Ethereum took an even bigger hit, crashing from $4,000 down to $3,500—a painful 10% reduction.

Massive Liquidations: Of the $708 million in liquidations, more than $600 million came from “long positions.” This means most traders were betting on prices to go up, and they lost big when the market turned against them.

Why is the Market in a Panic?

A perfect storm of global events and political uncertainty is driving this market chaos:

Global Turmoil: New, sweeping tariffs imposed by the White House on several countries have rattled financial markets worldwide. These trade tensions create a sense of fear, causing investors to pull money out of risky assets like crypto.

U.S. Jobs Report Controversy: A recent U.S. jobs report was so unprecedented that the official who released it was reportedly fired just hours later. This kind of political drama and economic uncertainty adds fuel to the fire, making investors nervous.

Geopolitical Tensions: Comments from President Trump about deploying nuclear submarines near Russian waters have also increased selling pressure, as geopolitical instability often makes investors cautious.

Experts Predict a Further Drop

According to market analysts, this could be just the beginning of the downturn. They are predicting that Bitcoin’s price could fall further to around $80,000 before it starts to rebound later in the year.

The “Fear & Greed Index” Signals Caution

A key tool used by crypto investors, the Crypto Fear & Greed Index, has dropped to 55. This number is very close to a “neutral” score of 50, which is a major shift from the recent optimism. It means that “fear” is starting to outweigh “greed” in the market, with investors now focusing more on protecting their money than on making big profits.

In short, the crypto market is in a period of significant adjustment. It’s being influenced not just by its own internal dynamics but also by the complex and unpredictable world of global politics and economics.

Join our WhatsApp community

Join Our Social Media Channels:

WhatsApp: NaijaEyes

Facebook: NaijaEyes

Twitter: NaijaEyes

Instagram: NaijaEyes

TikTok: NaijaEyes

READ THE LATEST TECH  NEWS

In a landmark moment for education on the African continent, the African Union Development Agency–NEPAD (AUDA‑NEPAD) has released the draft EdTech 2030 African plan for education technology Unveiled, a bold framework designed to revolutionise learning by 2030. Officially presented at the recent STEMtastic Adventures! The Africa symposium in Nairobi, this strategic blueprint lays out a clear roadmap for building a digitally‑inclusive education ecosystem across Africa.

A Vision Rooted in Continental Priorities

The EdTech 2030 African plan for education technology unveiled aligns closely with flagship continental initiatives such as Agenda 2063, STISA‑2034, CESA 2026‑2035, and the AU Digital Transformation Strategy. By integrating these pillars, AUDA‑NEPAD ensures that the framework supports broader ambitions for socio‑economic growth, innovation, and inclusive development.

Speaking at the unveiling, Dr. Barbara Glover, a key program officer at AUDA‑NEPAD, explained the intent behind the draft: “We are leveraging local innovation and leadership in EdTech to make Africa’s education systems more inclusive, resilient and future‑ready.” Her comments echo the belief that Africa today has the opportunity to leapfrog outdated educational systems through mobile‑first and offline‑capable digital learning solutions.

Africa’s Leapfrogging Moment: EdTech as Equaliser

Just as mobile money bypassed traditional banking infrastructure in Africa, the EdTech 2030 African plan for education technology unveiled positions digital learning as the continent’s leapfrogging opportunity in education. With unreliable legacy systems, many African countries can skip the traditional path and directly embrace scalable, technology‑driven education models ⎯ especially offline-first tools that work over low bandwidth or intermittent power.

Smartphone penetration provides a solid starting point: teacher smartphone ownership exceeds 90% in South Africa and ranges from 30–65% in Ghana, Nigeria, and Kenya. This existing infrastructure boosts feasibility for mobile‑based EdTech interventions that reach marginalized learners quickly and affordably.

Six Strategic Pillars of the EdTech Vision

At the core of the EdTech 2030 African plan for education technology unveiled are six strategic priorities, each crafted to ensure inclusive, scalable, and interoperable educational transformation:

1. Access & Infrastructure
   Expand connectivity with affordable devices, solar-powered solutions, and offline-first learning tools tailored for rural and underserved regions.
2. Curriculum-Aligned, Localised Content
   Promote homegrown, multilingual courseware that aligns with national curricula. Senegal’s XamXam app—delivering lessons in Wolof to over 1.2 million users—is highlighted as a success model.
3. Teacher Upskilling
   Equip educators with digital pedagogical skills, tools for content curation, and data literacy to integrate EdTech effectively into classrooms and remote learning environments.
4. Interoperability & Standards
   Implement vendor-neutral interoperable systems and open standards that enable seamless cross-border content sharing and platform integration.
5. Policy & Governance
   Harmonise policies across countries to support equitable funding, data privacy, inclusive regulation, and coordinated oversight.
6. Data & Research
   Establish evidence-based systems for monitoring learning outcomes, shaping responsive policy, and continuously improving EdTech interventions across the continent.

Phased Roll‑Out: Roadmap 2024–2030

The implementation of the EdTech 2030 African plan for education technology unveiled is structured into three mutually reinforcing phases:

• Phase I (2024–2026): Foundation Building
  Establish the technical, legal, and policy infrastructure. This includes setting interoperability standards, boosting cross-country collaboration, and clarifying governance frameworks.
• Phase II (2026–2028): System Integration
  Scale interoperable Digital Public Infrastructure (DPI), enable single‑sign‑on systems, and roll out regional courseware platforms and teacher support tools.
• Phase III (2029–2030): Consolidation & Global Leadership
  Position Africa as a global exporter of EdTech innovations. Establish a Pan‑African EdTech Innovation and Research Hub to support ongoing innovation, research, and export strategy.

Critical Imperatives: Context & Challenges

The plan responds to urgent education challenges. Sub-Saharan Africa has over 30 million children out of school, and by 2030, the continent will need 17 million new teachers to sustain universal access to quality education.

Only around 40% of African primary schools have internet access, while reports indicate up to 75% of youth lack critical digital skills needed for participation in the modern workforce. The youth cohort already accounts for over 60% of the continent’s population and is expected to reach 2.5 billion people by 2050.

By leveraging mobile-first EdTech and targeted infrastructure strategies, the plan seeks to level the playing field for learners in both urban and remote communities.

Stakeholder Roles & Collaborative Action

Sustainable success under the EdTech 2030 African plan for education technology, requires coordinated support across multiple actors:

• Governments: Align national education policies, curricula, and budgets with digital education ambitions.
• Regional Bodies: Oversee interoperability standards, research coordination, and shared platforms.
• Private Sector & Entrepreneurs: Provide low-cost devices, localised content, apps, and infrastructure solutions.
• NGOs & Development Partners (UNESCO, UNICEF, GPE): Contribute expertise, research capacity-building, and financial support.
• Local Communities: Champion awareness, inclusivity, and grassroots ownership of EdTech initiatives

Financial sustainability is envisioned through blended models: combining donor funding, private investment, licensing, and public‑private partnerships. This hybrid approach aims to fuel growth, scale promising solutions, and ensure affordability and access across diverse contexts.

Why This Matters: EdTech as the Engine of Transformation

For many African countries, traditional educational structures are struggling to meet demand. Digital solutions offer a potent alternative—compacting costs, delivering tailored content, and reaching learners outside formal school walls.

The EdTech 2030 African plan for education technology unveiled, aims to unlock that potential. Through standardised infrastructure, teacher development, and value-aligned policy, nations can build EdTech-powered education systems that are inclusive, scalable, and home‑grown.

Successful examples across the continent—like XamXam in Senegal or uLesson in Nigeria—demonstrate locally driven innovation at scale. Aligning those successes under a continental strategy could amplify impact, drive intra-Africa collaboration, and set Africa on a path to become a global EdTech exporter.

What Comes Next: Public Consultation and Inputs

AUDA‑NEPAD has invited stakeholders—governments, educators, tech creators, civil society, development agencies, and students—to review the draft. Written submissions are open until August 29, 2025, allowing feedback on how the plan could better reflect local realities and accelerate EdTech adoption continent-wide.

This consultation phase is pivotal. It will shape the final policy architecture, ensure inclusive participation, and set realistic commitments for implementation across nations.

Final Thoughts

The EdTech 2030 African plan for education technology unveiled represents a transformative moment in African education. Combining local innovation with continental scale, this strategy could pave the way for universal access to quality education, powered by digital tools, built on local content, and driven by inclusive governance.

If executed effectively, by 2030 Africa will not just close education gaps—it will lead a new era of globally competitive, locally relevant EdTech innovation. The future of education in Africa is digital, and this draft framework lays the groundwork for learning systems that are equitable, resilient, and visionary.

Join our WhatsApp community

Join Our Social Media Channels:

WhatsApp: NaijaEyes

Facebook: NaijaEyes

Twitter: NaijaEyes

Instagram: NaijaEyes

TikTok: NaijaEyes

READ THE LATEST TECH  NEWS

Artificial intelligence is swiftly transforming education worldwide—but amid this evolution, Africa stands at a critical crossroads. The continent cannot afford to be left behind. Moyo Orekoya, co‑founder of the AI‑powered teaching platform Shepherd, is sounding the alarm: Africa must seize this revolution, particularly within classrooms, or risk a widening educational and technological divide arXiv+5Punch+5Punch+5.

Shepherd: AI Tailored for African Classrooms

Shepherd is designed with African educators and learners in mind. Orekoya emphasises its ability to deliver personalised learning, adaptively guiding students based on their strengths and challenges. In environments where teacher quality and resources vary greatly, such tools offer vital support.

Personalisation is critical. Shepherd uses AI to assess student performance in real time, offering targeted feedback and scaffolding lessons to match individual pace. This significantly lightens the burden on overstretched teachers and enhances learning outcomes, in line with global research underscoring AI’s promise for individualised instruction in low-resource settings.

Tackling Real Classroom Challenges

Africa faces a persistent educational deficit: overcrowded classrooms, outdated curriculum delivery methods, and large student-to-teacher ratios. AI-driven tools like Shepherd can help address these systemic issues.

By automating routine assessment, Shepherd offers immediate insights into student progress. Teachers can then devote their limited time to designing richer lessons or offering one-on-one guidance. Shepherd also supports first‑language learning, inclusive education for learners with disabilities, and curriculum localisation—all key considerations in Africa’s diverse linguistic landscape.

Empowering Educators, Not Replacing Them

Orekoya is clear that AI isn’t a replacement for teachers—it’s a complement. Shepherd is built to empower educators, providing tools that make lesson planning more efficient and diagnostics more accurate.

This aligns with expert consensus that maximum learning gains occur when teachers are partners in the AI loop, not bypassed. When teachers are central to AI initiatives—from design to deployment—classroom adoption improves dramatically.

Evidence in Action: Early Wins

While Shepherd itself is early stage, similar AI‑based programs have delivered measurable results across Africa:

• In Ghana, the Rori chatbot tutor—accessible via WhatsApp—helped grade 3–9 students improve math performance (effect size ~0.37) over eight months, using low‑bandwidth devices.
• Kwame for Science, deployed across West Africa, served 750 users asking over 1,500 science questions. It showed top‑3 answer accuracy of 87.2%, demonstrating its ability to support students at scale.
• Broader projects supported by IDRC, AI4D, GPE‑KIX and national Ministries have applied AI to create textbooks in local languages, assist learners with disabilities, and train policymakers on AI competencies—highlighting systematic alignment with Africa’s Continental AI Strategy.

Together, these initiatives affirm that context-aware, ethically guided AI tools can be both effective and equitable in African educational contexts.

Toward a Responsible AI Agenda in African Education

Realising AI’s full potential in Africa requires infrastructure, policy, and human capacity. IDRC’s recent work emphasises:

• building AI competency frameworks for teachers and students,
• formulating national EdTech and AI policies,
• integrating AI literacy into teacher training,
• and promoting local, culturally-grounded innovations.

Orekoya and other innovators urge that Shepherd and similar platforms be developed responsibly, embedding ethics, data privacy, linguistic diversity, and accessibility from the ground up.

What Needs to Happen Now

To make AI in classrooms a widespread reality in Africa, several priorities must align:

1. Investment and Infrastructure

Reliable electricity, affordable devices, and stable internet access remain uneven across the continent. AI platforms like Shepherd must support offline or low-data modes. Public-private impact funding should accelerate infrastructure deployment across rural and underserved schools.

2. Teacher Training and Ownership

Teachers need training on AI tools—not just how to click buttons, but how to interpret data-driven insights and adapt teaching accordingly. Ongoing professional development and co-design processes improve trust and pedagogical integration.

3. Policy and Governance

National education authorities must create AI-friendly policies: clear standards for data protection, equity frameworks for classroom AI use, and guidelines ensuring alignment with national curricula and languages.

4. Localised, Scalable Innovation

Solutions like Shepherd must be grounded in African realities. Tools built for Ghana may not automatically work in Nigeria, Kenya, or Mali. Localisation—linguistic, cultural, curricular—is essential for legitimacy and impact.

5. Long-Term Impact Research

Sustained evaluation is key. Ghana’s Rori study showed promising short-term gains; Shepherd and other platforms should aim for longitudinal studies measuring retention, progression, and student resilience.

Why It Matters: Equity, Economy, and Future Readiness

Africa has missed previous industrial revolutions—web, broadband, smartphones —but AI presents a new inflection point. Access to AI-powered education can help:

• Close equity gaps between urban and rural, high- and low-income schools.
• Build digital literacy and problem-solving skills necessary for future jobs.
• Empower learners in local languages, helping preserve heritage while advancing education.
• Fuel economic development, with a more skilled workforce ready for data-intensive roles.

Orekoya believes Shepherd is part of this transformative generation of tools that can accelerate Africa’s leapfrogging into an AI-empowered future.

Classroom Voices: Teacher-Centred AI

While Shepherd is teacher-empowering by design, frontline educators report similar tools reduce admin load, help them identify struggling students earlier, and integrate interactive content more easily. Teachers appreciate having AI as an ally rather than overshadow.

This resonates with global findings: educators are more likely to adopt AI platforms when they retain pedagogical agency and can provide feedback to developers—a virtuous circle of improvement and teacher agency.

A Vision for Africa’s Classroom Future

Imagine a classroom in Nigeria, Ghana, Kenya, or Uganda where:

• A student logs into Shepherd on a basic tablet.
• Shepherd adapts the lesson to her current level—perhaps in math or science—providing dynamic exercises.
• The teacher receives an AI-generated report highlighting topic areas needing attention.
• Educators adjust group activities accordingly; brighter learners are challenged, and struggling learners get targeted support.
• The software works offline when electricity or the internet fails.
• Instructions, assessments, and feedback appear in local languages or dialects, increasing learner comfort and comprehension.

This isn’t a futuristic dream—it’s achievable today through tools like Shepherd, Rori, Kwame and companion innovations.

The Path Forward: Act Now

Africa’s educational systems are navigating heavy challenges: teacher shortages, outdated learning materials, motivation gaps, pandemic-related learning loss, and resource inequality.

AI-powered platforms offer a compelling, scalable solution—but only if implementation is intentional, inclusive, and collaborative. Shepherd co‑founder Orekoya urges urgency—not just in tech adoption, but with a vision: anchored in teacher empowerment, localised design, data privacy, and sustained investment.

Failing to act means putting an entire generation of young learners at risk of being left behind as the world embraces AI-augmented education.

Conclusion

Africa stands at the threshold of a transformative moment in education. Platforms like Shepherd, developed by local innovators, offer a hopeful blueprint for decades of opportunity to come. By aligning policy, investment, teacher training, infrastructure, and cultural localisation, African countries can—and must—ensure that their students are at the centre of the AI revolution in classrooms.

This is more than a technological shift—it’s a chance to write a new chapter in African education, one where learners everywhere have access to personalised, inclusive, and empowering instruction powered by AI.