The successive launches of chatbot DeepSeek, swiftly followed by autonomous agent Manus AI has meant that Chinese AI innovation has landed a one-two punch on existing Western foundational models. DeepSeek’s debut alone wiped a trillion dollars off the value of US stocks in the blink of an eye, while Manus AI’s launch also caused US […]
The successive launches of chatbot DeepSeek, swiftly followed by autonomous agent Manus AI has meant that Chinese AI innovation has landed a one-two punch on existing Western foundational models.
DeepSeek’s debut alone wiped a trillion dollars off the value of US stocks in the blink of an eye, while Manus AI’s launch also caused US tech stocks to tumble.
As well as unsettling US tech giants, these shock launches could be a pivotal moment for the AI industry as a whole – and a game-changer for the businesses queuing up to use it.
DeepSeek’s promise of cheap, easy-access AI calls into question the huge amounts of cash being splashed by market leaders like Nvidia, leading some to describe it as a ‘Sputnik moment’ – a disruptive, unexpected swoop into an arena that the US was getting comfortable in.
Just as businesses were digesting this disruption, another seismic shift arrived: the launch of Manus AI.
DeepSeek: Will leaner greener chatbot upend US Big Tech?
Developed by Chinese startup Monica, Manus AI is pushing the boundaries of autonomous intelligence, redefining how AI tackles real-world challenges. Unlike traditional, one-size-fits-all models, Manus operates through a sophisticated multi-agent architecture, where a network of specialised sub-agents—handling planning, knowledge retrieval, code generation, and more—work together.
By breaking down complex tasks into smaller, manageable parts and delegating them intelligently, Manus makes decisions and executes tasks with minimal human input. This game-changing approach paves the way for AI systems that are not just assistants, but fully independent problem solvers.
Manus AI builds on the trend DeepSeek helped expose—moving away from monolithic, cloud-reliant AI and towards leaner, more efficient, and cost-effective models.
As the rattled markets recovered composure, there was a rush to debunk DeepSeek’s financial claims with questions raised over security and harmful content. Meanwhile Manus AI’s autonomous capabilities pose key regulatory and ethical concerns.
Yet, whatever the issues with DeepSeek or Manus AI, there’s no arguing about their huge impact on the industry.
I predict the surprise emergence of these new players will usher in the following paradigm shifts:
-
A new era of AI design
The market-busting headline is that DeepSeek is cheaper than its US counterparts. But what we’re witnessing is a shift from the brute force computation being pushed by big tech and a move towards intelligent, optimised architectures. DeepSeek illustrates that the future of AI doesn’t need to be about throwing more expensive hardware at the problem – it’s about designing models that are smarter, faster and more economical.
Meanwhile, Manus AI’s multi-agent system allows businesses to tailor AI’s functionalities to their particular needs. By assigning specialised agents to handle distinct tasks, companies can develop bespoke AI solutions that integrate seamlessly into their operations, providing a strategic advantage over generic models.
-
The AI arms race is over; the future is specialisation
At the end of last year, Morgan Stanley predicted that the big tech giants would invest around $300bn in AI this year, but DeepSeek poses questions over whether that’s money well spent.
AI’s future isn’t about having the biggest model but the right model for your business. The likes of DeepSeek-R1 and Qwen2.5-Max show that targeted AI models can be just as powerful when fine-tuned for specific use cases.
Companies that customise AI will outcompete those relying on one-size-fits-all solutions which means that the age of generic AI is over.
-
By 2026, most businesses will have their own proprietary AI
It may sound a bold claim, but ‘build-your-own AI’ isn’t far off thanks to the shift towards small, specialised AI. Every major business could train its own model – the rise of scalable OpenSource AI models from China demonstrates that companies don’t need to rely on the bigger players or have billion-dollar budgets to build their own.
Start-ups and scale-ups can now launch AI-powered products without massive infrastructure costs, a development that levels the playing field with incumbents, potentially reshaping the competitive landscape.
Those experimenting now will gain a competitive advantage in automation, efficiency and AI-driven decision-making, without breaking the bank. Crucially, having their own models means companies avoid some of the security concerns that come with sending datasets out of house.
-
From consumers to creators
Manus AI is designed to integrate easily into business processes, with the aim of enhancing—not replacing—human expertise. This capability potentially allows companies building bespoke AI solutions to undergo a cultural shift from consumer to creator where employees actively shape AI-driven workflows rather than using ready-made tools.
Embracing this opportunity means a workforce more closely engaged in the design of AI-powered processes and workflows. Businesses that invest in AI literacy and upskilling staff to become creators will unlock new levels of productivity and innovation.
-
Risk teams will be a vital part of the AI ecosystem
As more and more companies realise that building their own AI models isn’t out of reach, the people involved in the process will create new workplace ecosystems. Some might be in-house, some not, but we’ll see ethics and risk management teams at the forefront: drawing up the company’s GenAI commandments, tackling GDPR and IP issues, and ensuring regulatory compliance.
And they will have their work cut out. As an autonomous AI agent, Manus AI raises considerable regulatory, ethical and compliance issues – what happens if it makes a mistake with serious financial or other harmful consequences?
Final thought: Startups, scale-ups and speed to market
DeepSeek and Manus AI aren’t just shaking up the AI industry—they’re redefining it. DeepSeek is showing that lean, cost effective AI can rival even the most powerful models, while Manus AI takes autonomy to the next level, turning AI from a tool into an independent problem-solver.
This shift gives businesses the option to be more than just AI consumers—they can become AI creators. Instead of relying on one size fits all models, they can now build custom AI solutions that drive competitive advantage.
But with autonomous AI comes risk. As Manus AI pushes the boundaries, companies must tackle regulation, accountability, and security head-on. The winners in this new era will be those who move safely and at speed, not just in adopting AI, but in making the technology a core, strategic asset.
