Reported by Musa Antiketu,| Journalist at Obaland Magazine
Anthropic chief executive officer, Dario Amodei, don call on companies wey dey build advanced artificial intelligence, AI, make dem slow down the speed wey dem dey improve their systems, as concern dey grow say technology fit reach level wey human beings no go fit control am properly.
Amodei make the call for one recent article, where e argue say AI development dey move very fast and say companies need to create more time to understand and manage the dangers wey fit come with more powerful AI systems.
According to Reuters report on Saturday, September 12, 2026, Amodei say AI companies need to deliberately reduce the speed wey dem dey increase the ability of their models. E add say even if the industry slow down, progress for AI still go look very fast, but the extra time fit help researchers, governments and technology companies prepare better safety measures.
The warning come at a time when competition among major AI companies don become stronger, with firms dey spend huge money to develop systems wey fit reason, write, code, conduct research and perform tasks with less direct human supervision.
The main concern be say as AI systems become more capable and independent, people fit find am harder to understand exactly wetin the systems dey do, monitor their actions or stop them when something go wrong.
Amodei’s warning also come shortly after Anthropic disclose cases where its Claude AI systems allegedly get used by different actors for activities connected to cyber operations, surveillance, fraud and research wey fit have dangerous biological applications.
Anthropic say some people try use its AI systems for harmful purposes, while the company put additional safeguards in place to reduce the possibility of serious misuse.
The development don add to wider argument inside the technology industry about whether AI companies should continue to compete at the current speed or agree on stronger limits and safety conditions.
Reuters recently report say some researchers inside major AI companies, including OpenAI and Anthropic, don raise concerns about the speed of development. Some of dem believe say leading AI laboratories need to consider a coordinated slowdown if safety measures no dey keep pace with the growing ability of the systems.
One Anthropic researcher, Jacob Coxon, also resign from the company this week and raise concern about what e describe as irresponsible competition toward highly advanced and potentially self-improving AI systems. Associated Press report say Coxon warned that poorly controlled AI development fit create serious danger if the technology eventually become much more powerful than human beings fit safely manage.
The debate no dey happen only among technology companies. Governments also dey look for ways to control the possible dangers connected to advanced AI.
For United States, lawmakers dey discuss proposals wey fit place stronger responsibility on AI developers to identify and reduce major risks before dem release highly capable systems. Reuters report say negotiators for the US Senate dey consider a possible legal framework wey fit require AI companies to take greater responsibility for preventing catastrophic risks.
At the same time, United States and China dey prepare for discussions about AI safety, as both countries face concern over cyberattacks, autonomous systems and other possible consequences of increasingly powerful AI technology.
For Africa, the debate get particular importance because AI adoption dey increase across education, banking, health, government services, media, business and communication. More advanced AI systems fit create new opportunities for African countries, but weak regulation, limited technical capacity and poor digital-security infrastructure fit also expose individuals and institutions to new forms of risk.
The issue therefore no simply be whether AI development should stop. The bigger question be how the world fit make sure say the technology continue to bring benefits while governments, researchers and companies still get enough ability to understand, monitor and control the risks.
Amodei’s position show the growing disagreement between the economic pressure to build more powerful AI systems quickly and the need to make sure say safety measures dey move at the same speed.
As the global AI competition continue, the question wey dey before technology companies and governments now be whether progress must always mean moving faster, or whether responsible progress sometimes require people to slow down so dem fit understand exactly where the technology dey go.
Sources
Reuters
Associated Press
Anthropic boss Dario Amodei don call on AI companies make dem slow down development as concern dey grow over human control, safety and misuse of advanced artificial intelligence.