Artificial intelligence is entering a new phase in which systems will not only generate content or analyse information but also execute financial decisions, manage digital assets and participate directly in economic activity, according to technology and finance executives marking AI Appreciation Day.
The transition is already supported by rapidly improving technical capabilities and growing investment.
Stanford University’s 2026 AI Index found that AI agents’ success rate on real-world tasks rose from 20% in 2025 to 77.3% in 2026. Global corporate AI investment reached $581.7 billion last year, an increase of 130% from 2024.
However, the report also found that current systems continue to struggle with multistep planning and financial analysis, highlighting the gap between experimental agents and dependable economic infrastructure.
AI Agents Begin Moving Beyond Assistance
That gap is beginning to narrow inside companies that are deploying agents across development, compliance and financial operations.
Ryan Kirkley, CEO and co-founder of Global Settlement Network, said he has watched AI move from an experimental technology into a central component of business operations.
“Having invested in AI companies for several years, I've had the chance to watch the technology move from an interesting experiment to something that is genuinely reshaping how businesses operate,” Kirkley told Yellow.com.
At Global Settlement, he said AI has become central to the company’s operations, with more agents than employees and agentic systems supporting compliance, identity and software development.
Kirkley sees the combination of AI and blockchain as particularly significant. AI systems can process data and identify opportunities, while blockchain networks provide programmable infrastructure through which financial decisions can be executed across borders.
“What I find most exciting is the intersection of AI and crypto because each technology unlocks something the other has been missing,” he said.
The Bank for International Settlements has similarly identified AI and tokenization as technologies that could bring trading, settlement and collateral management closer together while reducing reconciliation costs. The BIS said AI is already being used by financial institutions for credit underwriting, fraud detection, risk management and back-office automation.
Jordi Esturi, chief marketing officer at tokenization platform Brickken, said the industry has focused too heavily on current applications such as text generation, meeting summaries and coding assistance.
“The next frontier of AI is becoming an active actor in the economy, helping people execute financial decisions, manage digital assets and coordinate increasingly complex transactions in real time,” Esturi said.
He described this development as the basis of agentic finance and agentic capital markets, where AI systems operate within defined governance structures to support capital formation and asset management.
Under that model, founders could use AI-supported infrastructure to raise funds, investors could manage portfolios and companies could issue tokenised assets with fewer manual processes.
“A founder raising capital, an investor managing a portfolio or a company issuing tokenized assets should be able to interact with financial infrastructure as naturally as they use the internet today,” Esturi said.
The BIS said tokenized ledgers can support automated, round-the-clock operations and simultaneous settlement, although it warned that reliable money, clear governance and regulatory safeguards are necessary for such systems to operate at scale.
Also Read: RWA Tokenization Tripled But 80% Of Value Sits In Just One Asset Class
Emerging Markets Could Gain A Larger Role
Beyond finance, falling AI development costs are changing where technology companies can be built.
Lily Dash, co-founder of Actai Advisors and founder of Future Caribbean, said access to AI is weakening the historical connection between geography and participation in the global technology economy.
“For the first time, geography matters far less than talent, ambition and access to the right tools,” Dash said.
She pointed to Barbados, Jamaica, Trinidad, Nigeria and Kenya as markets where founders can now build products and contribute to AI development without relocating to established technology centres such as Silicon Valley or London.
Dash said the cost of participating in technology development has fallen sharply, giving regions that traditionally consumed imported technology a greater opportunity to produce and export their own products.
Stanford’s AI Index found that generative AI reached 53% population adoption within three years, faster than either the personal computer or the internet. Adoption still correlates strongly with national income, however, showing that access remains uneven despite the technology’s rapid spread.
The World Bank has also warned that low- and middle-income countries face substantial barriers to deploying AI at scale. Its Digital Progress and Trends Report identifies four foundations for wider adoption: connectivity, computing capacity, locally relevant data and workforce skills.
Dash said those foundations will determine whether emerging markets capture lasting economic value from AI.
“We have to make sure people have access to the infrastructure, education, mentors and investment they need to turn ideas into real businesses,” she said.
She argued that investment in local entrepreneurs could allow AI to support gross domestic product growth, high-value employment and a more geographically diverse generation of technology companies.
Governance Will Determine the Outcome
The executives broadly agreed that AI’s economic importance will increasingly come from what systems can do, rather than what they can generate.
Kirkley expects AI to make tokenised assets and digital money easier for businesses and consumers to navigate, potentially accelerating adoption of blockchain-based financial services.
“The future of finance won't just be digital, it will be intelligent by default,” he said.
Yet increased autonomy also introduces risks. The BIS has warned that similar AI models may lead financial institutions to respond to market shocks in the same way, amplifying volatility and liquidity pressure. Dependence on a small number of cloud, data and model providers could also create operational vulnerabilities.
Read Next: Perp DEXs Are Eating Centralized Volume, And The Shift Is Accelerating