"Every major shift in technology does two things: it destroys the familiar, and it creates something we couldn't have imagined before. The question is always which side of that are you on?"
We are living through one of those shifts right now.
Not the chatbot moment, that already happened. The shift underway in 2026 is quieter, more consequential, and far less understood by the people it will affect most. It is the move from AI that responds to AI that acts. From a tool you prompt, to a system that takes an objective, breaks it into steps, navigates multiple platforms, checks its own work, and delivers an outcome without asking for your help at every stage.
The word the industry has settled on is agentic AI. And whether you are a fresh graduate in Uyo trying to figure out where the jobs are, a fintech operator in Lagos watching your compliance costs, or an investor trying to read where capital is flowing this shift is the most important thing happening in your field right now.
Here is the clearest way to understand the difference.
A chatbot responds to your prompt. You type, it answers, the exchange ends. An agent is given a goal. It plans the steps to reach it, calls the tools it needs, moves across systems, and revises its approach based on what it finds, with limited human input. The distinction sounds technical. The consequences are not.
A chatbot that summarises a compliance report saves you twenty minutes. An agent that reads every trade log, cross-references regulatory requirements, flags anomalies, and drafts a remediation plan replaces a team of junior analysts.
That is not a hypothetical. Goldman Sachs is deploying autonomous AI agents built with Anthropic's Claude model to automate core accounting, compliance, and operational finance functions. The bank's chief information officer Marco Argenti told CNBC that Goldman had spent six months embedding Anthropic engineers within its technology teams to co-develop agents capable of handling transaction reconciliation, trade accounting, client vetting, and onboarding work that has resisted full automation for decades because it requires processing large volumes of data against strict regulatory frameworks.
Goldman Sachs Lets AI Agents Do Accounting and Compliance Work | PYMNTS.com
Goldman Sachs is deploying autonomous artificial intelligence (AI) agents built with Anthropic's Claude model to automate core accounting, compliance and
These are not the easy jobs. They are among the most rule-dense, liability-heavy workflows in institutional finance. The fact that agents are being deployed there rather than starting with simpler tasks is the signal that the technology has cleared a real threshold.
Goldman's CIO put the direction plainly: companies will shift from deploying human centric staff to tackle tasks, to deploying human orchestrated fleets of specialised multi agent teams.
What to Expect From AI in 2026: Personal Agents, Mega Alliances, and the Gigawatt Ceiling
AI models have reached an important inflection point, which will change how humans and machines work together.
That sentence is worth reading twice, because it describes not a future state but what Goldman is already building today.
Before going further, one figure needs to be on the table because every serious analysis of agentic AI has to start here.
Carnegie Mellon researchers, working with Salesforce, built a simulated workplace benchmark called The
Agent Company and found that the best-performing AI agent only completed about 24% of real-world office tasks end-to-end. Failure rates ran close to 70%, with agents struggling most on multi-step processes, ambiguous instructions, and tasks requiring social judgment.
What Is Agentic AI? A Guide for Businesses (2026) | SLIDEFACTORY
Agentic AI is the most talked-about technology in business right now. Here's what it actually does, where it works, where it doesn't, and how to know if your business is ready - from Portland's AI consulting team.
24% completion. 70% failure rate. Those are the current production numbers for the most capable agents on structured office tasks.
This is not an argument that the technology is overhyped. It is an argument for precision about where it works and where it does not. Agents are reliable today in narrow, well-defined, rule-heavy workflows, transaction processing, document classification, compliance checking, data reconciliation. Outside that lane, they break. The implication is that the disruption happening right now is concentrated in specific roles, not spread evenly across the economy. And those specific roles happen to be the ones that absorb the most new graduate.
Interactive version: https://medtee360.github.io/logliq-data/
The aggregate employment debate, will AI create more jobs than it destroys? is real and genuinely unresolved. But it misses the most important near term consequence of the agentic shift, which is not about total job numbers. It is about which jobs are disappearing and when.
The problem is not that AI will trigger mass layoffs. It is being framed incorrectly. The changes are already happening quietly. Many firms are not laying off their current employees at scale. Instead, they are silently closing the door to new ones.
AI won't kill your job - it will kill the path to your first one | Fortune
The AI disruption story is being told wrong. Firms aren't firing - they're freezing hiring.
That is the junior gap. Entry-level roles, the positions that absorb fresh graduates, build the first layer of professional experience, and create the tacit knowledge that makes experienced workers valuable later are being automated before the people who would have filled them ever get the chance to start. In the United States, Goldman Sachs reported in April 2026 that AI is erasing roughly 16,000 net jobs per month. Administrative roles face 26% direct risk. Customer service roles face 20%. In tech specifically, junior developer and QA roles are declining 20–35% globally while senior engineer and AI specialist roles are growing faster than companies can fill them.
AI Impact on the Job Market in 2026: What the Data Shows | Second Talent
AI is reshaping the 2026 job market: where roles are disappearing, where new ones are emerging, and what skills employers actually want now.
The global net picture by 2030 may well be positive. The transition is not. And in Nigeria, that transition hits a labour market with far less room to absorb the disruption.
The number that gets quoted about Africa's AI exposure 0.4% of employment directly at risk, compared to 5.5% in high-income countries is accurate and misleading in the same breath. It is an economy wide average being pulled down by the 70%+ of Nigeria's workforce in agriculture and the informal sector, which AI does not yet meaningfully touch at scale.
The urban, formal, graduate entry layer tells a completely different story.
If there is one sector in Nigeria where AI's workforce impact is already visible and measurable, it is financial services. By early 2024, thirteen Deposit Money Banks had integrated AI-powered chatbots UBA's "LEO," Zenith Bank's "ZiVA," and others handling queries that previously required call centre agents. These systems do not sleep, do not require shift allowances, and are available across every digital channel simultaneously. Customer service and branch facing roles have historically been among the largest entry-level hiring pools in Nigeria's formal banking sector: positions that absorb fresh graduates and serve as the gateway into corporate careers.
AI and the Nigerian Workforce: Which Jobs Are at Risk and How Employers Should Prepare - MacTay
There is a conversation happening in boardrooms, university lecture halls, government ministries, and even over jollof rice at family gatherings; it is,
As agentic AI matures, that pressure moves further up the chain from tier-1 customer service into analyst, administrator, and compliance roles. The same workflow that Goldman Sachs is automating in New York is the one that Nigerian banks will automate in Lagos. The technology does not respect geography. It follows the structure of the work.
But here is what distinguishes the Nigerian story from a simple import of Western disruption: Nigerian organisations are gaining hands-on AI experience that is enabling local companies to develop skill sets explicitly tailored to the Nigerian market accounting for infrastructure challenges, diverse languages, and local consumer behaviours. The observation from practitioners is that Nigeria is not just using AI; it is building a version of it that works here.
The future of work in Nigeria: How AI is redefining careers - Businessday NG
Nigeria's relationship with AI is unique. Unlike markets that are purely importing software, Nigerian organisations have already gained...
The evidence is visible in the job market right now. Flutterwave is actively hiring AI engineers to build production grade systems inside its payments infrastructure. M-KOPA has created a brand new AI Operations team to design and scale agents across the business. Dragnet Solutions is hiring an AI Enablement and Agentic Automation Lead. These are not research roles. They are operational hires, indicating that the build phase not just the adoption phase is underway inside Nigerian companies.
In 2026, NLP, MLOps, and AI agent skills are in very short supply in Nigeria, and engineers who carry them earn 30 to 50% more than general ML engineers at the same experience level. Tier 1 Nigerian banks and fintechs are paying ₦600,000 to ₦2.5 million per month for mid to-senior AI and data roles. The premium is real, the demand is real, and the pipeline to fill it remains thin.
AI Salary in Nigeria: Your Complete 2026 Guide
What does AI pay in Nigeria in 2026? Abuja Data School reveals every AI salary band from junior analyst to senior ML engineer. Earn more today.
The International Finance Corporation projects that 28 million jobs in Nigeria and 230 million across Sub-Saharan Africa will require digital skills by 2030. That is not an opportunity statement in isolation. It is a deadline with consequences for anyone who does not treat it as one.
Agentic AI needs infrastructure to be genuinely autonomous. An agent that can plan and reason but cannot settle, verify, or record transactions without human sign-off is still dependent on human intermediaries at the most critical moments. This is where Real-World Asset tokenisation enters the story not as a separate trend, but as the settlement layer the agentic economy is being built on top of.
According to data from rwa.xyz and PYMNTS reporting in March 2026, the on-chain RWA market excluding stable coins reached approximately $26 billion, up from roughly $6.5 billion a year earlier, a growth rate of around 300%. That is not speculative momentum. The passage of the GENIUS Act in July 2025 established a federal framework and standardised settlement infrastructure for payment stablecoins, triggering a surge of institutional capital. The transparency of public ledgers, real-time monitoring of asset flows and counterparties aligns precisely with institutional compliance requirements.
Tokenized RWAs and On-Chain Commodities
How the exponential growth of tokenized real-world assets (RWAs) and on-chain commodities is impacting global finance.
The market has moved from being dominated by a single asset class tokenised US Treasuries to having at least six categories that each independently exceed a billion dollars in on-chain value. That diversification changes the resilience of the sector: a market built on one asset class is one regulatory decision away from a major drawdown; a market built on six is structurally harder to dislodge.
For Nigeria, this infrastructure story has a direct local entry point. The CBN launched a crypto AML/CFT pilot in April 2026 with Flutterwave, Paystack, and Juicyway as the first cohort, requiring monthly performance metrics, direct supervisory engagement, and implementation plans for the FATF Travel Rule. That regulatory maturity is a prerequisite for Nigerian fintechs to participate credibly in the global RWA infrastructure being built and for Nigerian investors to access tokenised yield products that were previously only reachable through foreign platforms.
CBN Launches Crypto AML Pilot With Flutterwave, Paystack and Other VASPs
The Central Bank of Nigeria (CBN) has kicked off a new supervision pilot for virtual asset service providers (VASPs), tightening its oversight of
The direction of the RWA market is reliable. The scale of the 2030 forecasts ranging from McKinsey's conservative $2 trillion to BCG and Standard Chartered's $16–30 trillion reflects genuine uncertainty about the pace of regulatory harmonisation and institutional adoption, not about the direction. Treat the range as scenarios. Treat the trajectory as settled.
Three ideas that will make everything above click, in plain terms.
Agentic AI vs chatbot AI. A chatbot answers your question and stops. An agent is given a goal and figures out the steps to reach it on its own calling tools, reading data, and checking its own outputs along the way. The difference matters because agents can complete entire workflows, not just single tasks. That is why they threaten entire role categories, not just individual tasks within roles.
The junior gap. AI is not replacing experienced workers at scale right now. It is preventing the creation of the entry-level roles that would have become experienced workers. If the on-ramp to a career disappears, the career trajectory changes for everyone behind it. This is the most underreported consequence of the agentic shift and it hits a young, urban, formally-employed Nigerian workforce directly.
On-chain settlement. When a financial transaction is settled on-chain, it means the record of that transaction is written to a public blockchain - permanent, verifiable by anyone, and executed automatically when the conditions are met. This is what makes truly autonomous agents possible: they can act, settle, and record without a human approving each step. The RWA market is building the assets. Stable coins are building the cash layer. Agents are the actors that will move between them.
The agentic shift is not arriving. Goldman Sachs is already in production. Nigerian banks deployed their first AI agents two years ago. The CBN is building the regulatory infrastructure for the next layer. The RWA market has grown 300% in twelve months.
What is not yet settled is who benefits from this in Nigeria. The infrastructure is being built regardless of whether the workforce is ready for it. The African Development Bank projects that inclusive AI deployment could contribute up to $1 trillion in additional GDP across Africa by 2035 — if the continent strengthens its digital foundations and workforce capabilities. That conditional is doing enormous work. The trillion dollars is the upside of preparation. The downside of the same transition, for a workforce that does not adapt, is a junior gap that closes the door on an entire generation of formal employment.
From demographic dividend to digital power: AI and the future of work in Africa
Africa's future of work will be shaped by AI, digital skills and workforce readiness. With millions of young people entering the labour market each year, closing the skills gap is now essential to productivity, competitiveness, job creation and inclusive long-term growth across the continent.
The logic is clear. Whether the liquidity follows depends on decisions being made right now, in boardrooms, universities, and CBN policy offices.
Logic makes the plan. Liquidity makes it grow.
Logic & Liquidity — Finance. Tech. Web3. Business.