Latest News
OpenAI CEO Sam Altman projects USD 20 billion revenue, backs US chip push Microsoft gains OpenAI deal allowing independent AGI pursuit or partnerships AI drives three initiatives to accelerate global action for planet protection Gemini API adds File Search Tool supporting PDF, DOCX, TXT, JSON, code files BBVA scales AI to 11,000 staff, saving 3 hrs weekly per employee Moonshot's K2 Thinking tops open‑source AI, beating GPT‑5 and MiniMax‑M2 AI expected to make small discoveries by 2026, larger ones from 2028 onward Anthropic's Claude reduces SOC investigations from 5 hrs to 7 min, 95% accuracy OpenAI CEO Sam Altman projects USD 20 billion revenue, backs US chip push Microsoft gains OpenAI deal allowing independent AGI pursuit or partnerships AI drives three initiatives to accelerate global action for planet protection Gemini API adds File Search Tool supporting PDF, DOCX, TXT, JSON, code files BBVA scales AI to 11,000 staff, saving 3 hrs weekly per employee Moonshot's K2 Thinking tops open‑source AI, beating GPT‑5 and MiniMax‑M2 AI expected to make small discoveries by 2026, larger ones from 2028 onward Anthropic's Claude reduces SOC investigations from 5 hrs to 7 min, 95% accuracy
Our content generation service is experiencing issues. A human-curated summary is being prepared.
Business & Startups

BBVA scales AI to 11,000 staff, saving 3 hrs weekly per employee

By
• 3 min read

BBVA’s internal AI rollout reads more like a logistics puzzle than a tech showcase. While the bank experimented with a handful of custom GPTs in isolated departments, the real test came when those tools moved beyond pilot labs and into everyday workflows. Over the past year, the lender reported that roughly 20,000 bespoke models were built, yet only about 4,000 see regular use.

When those frequently accessed bots were put through workflow simulations, some processes sped up by more than 80 percent. Employees who adopt the assistants clock an average of three saved hours each week, and weekly active usage now sits at 83 percent across the organization. The numbers suggest a shift from novelty to utility, but scaling any AI effort carries operational risk.

How did BBVA bridge the gap from a few thousand early adopters to a bank‑wide deployment? The answer comes straight from the team that managed the expansion.

"We deployed initially to 3,000 employees, then very quickly jumped to 11,000 employees." - ~3 hours saved per employee per week - 83% weekly active usage - Efficiency improvements up to 80%+ in workflow tests - 20,000+ Custom GPTs created across the bank, with around 4,000 GPTs used frequently BBVA built its AI program on trust, governance, and structured learning--enabling scale without compromising security. "We created this atmosphere of being in a safe place to learn and to use AI," says Alfaro. Rather than pushing experimentation underground, BBVA brought it into a trusted environment and reduced shadow-AI risk by giving employees secure tools and clear guardrails. "Instead of shadow AI… we gave them a platform that was safe so they could start experimenting," Alfaro notes, "and we gave specific training for 250 leaders, including our CEO and chairman." - ChatGPT Enterprise deployed across teams globally - Security, legal, and compliance aligned from the start - Thousands of Custom GPTs created by employees - 250 senior leaders trained, including the CEO and chairman This structured approach is now showing results at scale.

Related Topics: #AI #BBVA #custom GPTs #ChatGPT Enterprise #workflow #security #governance #shadow-AI

BBVA’s rollout shows AI moving beyond pilots. After an initial deployment to 3,000 staff, the bank quickly expanded the tools to 11,000 employees. Each user reportedly saves about three hours per week, and weekly active usage sits at 83 percent.

In internal workflow tests, efficiency gains topped 80 percent, and more than 20,000 custom GPTs have been built, with roughly 4,000 seeing regular use. The numbers suggest a tangible shift in how routine tasks are handled. Yet the broader impact on profitability or customer outcomes is not yet disclosed.

How the bank will integrate these models with legacy systems remains uncertain, and whether the productivity gains will scale as the user base grows is unclear. The initiative reflects BBVA’s intent to embed AI as a core capability rather than a side project. Still, the long‑term sustainability of such high usage rates will need monitoring.

For now, the data points to a significant internal efficiency push, but external validation is still pending.

Further Reading

Related Articles

Common Questions Answered

How many BBVA employees are currently using the internal AI tools, and what was the initial deployment size?

BBVA expanded its AI rollout from an initial 3,000 employees to 11,000 staff members. This rapid scaling means that roughly one‑third of the bank’s workforce now regularly accesses the custom GPT tools.

What time savings does BBVA report per employee after adopting the AI tools?

Each employee using BBVA’s AI suite saves about three hours per week, according to the bank’s internal metrics. This reduction translates into significant productivity gains across routine tasks and processes.

How many custom GPT models has BBVA created, and how many are used frequently?

The bank has built more than 20,000 bespoke GPT models across its divisions, but only around 4,000 of those see regular, frequent use. The disparity highlights the experimental nature of many models while focusing resources on the most valuable bots.

What efficiency improvements have BBVA’s workflow simulations shown after AI integration?

Internal workflow tests indicate efficiency gains of up to 80 percent when AI‑driven bots are employed. These improvements, combined with an 83 percent weekly active usage rate, demonstrate the tangible impact of the AI rollout on operational speed.