Artificial Intelligence in full swing: OpenAI, Google AI, ChatGPT, and the moves reshaping the industry
Artificial Intelligence has never been this hectic.
Between team expansions, new models being launched, and growing questions about transparency in the industry, what we’re seeing is an ecosystem that just won’t stop reinventing itself.
And at the center of it all, names like OpenAI, ChatGPT, and Google AI keep setting the pace for change. 🚀
But what really stands out right now isn’t just the speed of technological advances. It’s the way everyone is adapting to them — from major corporations to researchers who don’t even write a single line of code anymore.
In this latest round of news, there’s a lot going on:
- OpenAI planning to nearly double its team by the end of 2026, reaching 8,000 employees
- Indian companies leading global AI adoption but still stumbling over a lack of expertise
- A serious debate about transparency following the Cursor and Kimi model situation
- Andrej Karpathy revealing he no longer writes code directly and spends hours steering AI agents instead
- OpenAI gearing up to expand ads in ChatGPT for all free users and those on lower-cost plans
And that’s not even counting the infrastructure moves, like Elon Musk’s Terafab project developing chips to operate in space through Tesla and SpaceX, security concerns that reached the Pentagon involving Anthropic, and an Indian company building a firewall specifically for AI actions. 🛰️
The word that ties all of this together is simple: adaptation. And that’s exactly what we’re going to talk about here.
OpenAI in expansion mode: what does nearly doubling in size by 2026 mean?
According to the Financial Times, OpenAI plans to nearly double its headcount to around 8,000 people by the end of 2026. Most of the new hires will be directed toward product development, engineering, research, and sales. The company is also ramping up recruitment of specialists focused on what they call technical ambassadorship, aimed at helping businesses make better use of OpenAI tools.
When a company plans to grow this aggressively in such a short time frame, it’s not just another corporate headline. OpenAI is signaling something much deeper: the race for artificial intelligence talent has entered a new phase, and anyone standing still is going to lose ground fast. The plan shows that the organization wants to maintain the breakneck development pace of its models, including upcoming versions of ChatGPT, while simultaneously expanding its footprint in areas like safety, fundamental research, and technical support for enterprise clients.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
This move also has a direct impact on the global job market. Engineers, researchers, systems designers, and AI ethics specialists are being fought over like never before. Smaller companies feel this effect firsthand because top talent tends to gravitate toward the conditions that giants like OpenAI can offer. But the interesting part is that this growth also opens doors for people just starting out, since demand for qualified professionals is so high that the industry needs to invest in training and internal development with an urgency that simply didn’t exist before.
And alongside this team expansion comes another important development: OpenAI plans to expand ads in ChatGPT for all free users and lower-cost plans, according to The Information. This shows the company is diversifying its revenue streams to sustain this growth pace — a clear sign that monetization of language models is entering a more mature and aggressive phase.
The transparency debate nobody can ignore
The situation involving the code editor Cursor and the Kimi model set off an alarm that a good chunk of the artificial intelligence community had been waiting for. Users speculated that Composer 2, a new model designed to improve efficiency in software development workflows, had been built on top of an external base model that wasn’t disclosed at launch. In a post on X, Cursor co-founder Aman Sanger acknowledged that they did fail to mention the Kimi base in the launch blog post and stated the company would correct this in future releases.
The situation exposed an issue that stays in the background when everything is working fine but becomes a serious problem when something goes wrong: to what extent do users actually know which model they’re interacting with? When a tool swaps or hides the underlying model without telling anyone, that’s not just a technical issue. It’s a trust issue — and trust is exactly the hardest asset to recover once it’s lost.
This episode sparked a much broader discussion about how companies like Google AI, OpenAI, and others in the space should communicate changes to their models in a clear and accessible way. The lack of standardization around what needs to be disclosed and when creates a gray area that can be exploited in ways that hurt the end user. And the problem is that the average user relying on ChatGPT or any other AI tool day to day rarely has the technical know-how to notice when something changed under the hood. They just notice the experience feels different but don’t know why. 🔍
Transparency in AI isn’t just a talking point for activists or academics. It has very real, practical consequences — including for national security, which explains why the topic has reached Pentagon-level discussions. Anthropic, for example, recently challenged a Pentagon claim about national security risks in response to a lawsuit, showing that the debate around transparency and accountability in AI is becoming increasingly institutional and legal.
Andrej Karpathy and the new way of working with AI
Few people in the world have the technical credibility of Andrej Karpathy to talk about how artificial intelligence is changing the work of those who build technology. A former director of AI at Tesla and one of the most respected names in large language models, Karpathy revealed in an interview on the No Priors podcast that he no longer writes code directly. Instead, he spends long hours expressing intentions to AI systems, steering agents rather than programming manually.
Karpathy described his current state as a kind of AI psychosis — referring to an intense focus on using artificial intelligence tools and the belief that the rapidly expanding capabilities of these tools make almost anything possible. According to him, the main bottleneck today is no longer computational power but the human ability to effectively direct AI systems.
What makes this revelation especially interesting is what it implies for the future of tech-related careers. If even the best engineers in the world are redefining what it means to work in AI, the rest of the market is going to need to do the same — and faster than they think. This isn’t about people being replaced by machines, at least not in the simplistic way that debate usually gets framed. It’s about a real shift in the skill set that matters, where reasoning ability, critical judgment, and contextual understanding are worth more than the ability to memorize programming language syntax.
Professionals who know how to work alongside Google AI, OpenAI, and similar tools productively and critically are becoming the new standard of excellence in the industry. It’s not enough to know how to use ChatGPT for simple tasks. The real differentiator is knowing when to trust the AI, when to question its output, and how to integrate that workflow so the final result is better than either one — human or machine — could achieve alone. Karpathy is basically demonstrating in real time what it means to work well in the age of artificial intelligence. 🤖
Emerging markets in the AI race: the case of India
The adoption of artificial intelligence by Indian companies has drawn attention in recent global surveys, and for a very telling reason. According to a Deloitte report, Indian companies are adopting AI at scale, outpacing global competitors. Most organizations plan to increase AI spending in the coming year. However, that rapid deployment stands in stark contrast to a shortage of artificial intelligence expertise within those very same companies.
Despite the challenges, Indian companies expect significant productivity gains from AI. The focus is shifting from experimentation to effectively embedding artificial intelligence into value creation and competitive advantage. But the gap between adoption and understanding is one of the most complex challenges the global ecosystem will need to tackle in the years ahead.
There’s no point in releasing increasingly powerful models if the knowledge infrastructure to use them well can’t keep up. And we’re not just talking about technical training here. We’re talking about a deeper cultural shift in how organizations make decisions, how they measure results, and how they define success when an AI tool is part of the process. That kind of transformation takes time, and compressing it too much can create more problems than solutions.
Still on the Indian front, it’s worth noting that a Gujarat-based company developed a system called AI Action Firewall, designed to make artificial intelligence systems safer. This shows that concern about AI safety isn’t exclusive to big American or European corporations. Emerging markets are building their own solutions to deal with the risks that accelerated adoption brings along.
What makes the Indian landscape relevant for everyone is that it works as a large-scale lab for how AI adaptation happens in practice — with all the contradictions that entails. Companies moving fast, talent in short supply, regulation still in its early stages, and enormous pressure to show results. 🌏
Invisible infrastructure: the Terafab project and chips in space
Behind every conversation with ChatGPT or every search powered by Google AI, there’s a layer of infrastructure that rarely makes headlines but determines what’s actually possible with artificial intelligence in the real world.
The Terafab project, launched by Elon Musk’s Tesla and SpaceX, envisions creating space-based AI computing using solar-powered satellites that could host data centers in orbit. The project starts with 100-kilowatt capacity and is expected to scale to megawatt levels, leveraging the constant solar energy available in space to cut costs compared to ground-based systems.
Two types of chips are being planned: one for terrestrial applications like Tesla’s vehicles and robotics systems, and another called D3, designed specifically for space environments. Satellites with onboard AI processing capability open up enormous possibilities for environmental monitoring, communications, autonomous navigation, and a range of other applications that depend on real-time processing in places where you can’t rely on connections to servers on the ground.
And when it comes to strategic AI applications, the conversation inevitably reaches the government level. The fact that security concerns related to artificial intelligence are being actively discussed — with companies like Anthropic legally challenging Pentagon claims about national security risks — is an important indicator that the experimentation phase is fading. AI is being woven into critical decision-making processes, and that demands a level of robustness, auditability, and control that current models are still building toward.
Elon Musk and Demis Hassabis: AI and scientific discovery
Alongside the infrastructure moves, it’s worth noting that Elon Musk and Demis Hassabis, CEO of Google DeepMind, publicly exchanged their views on the role of AI in scientific discovery. This kind of conversation between figures of such influence in the industry helps shape the public narrative around where artificial intelligence is headed and which applications should be prioritized.
Hassabis has consistently championed the use of AI as a tool for accelerating discoveries in fields like biology, chemistry, and physics, while Musk tends to emphasize existential risks and the need for stricter controls. The dialogue between these two perspectives is healthy for the ecosystem because it keeps the industry from falling into a one-sided view — whether overly optimistic or overly pessimistic.
What all of this means in practice
All of this together — from OpenAI‘s expansion to the Terafab project’s orbital chips, from the transparency demanded in the Cursor case to the new way of working described by Karpathy — forms a coherent picture of an industry growing in every dimension at once. Computational capacity, geographic reach, diversity of applications, and complexity of challenges are all scaling together.
AI-powered WealthTech is also gaining traction, with venture capitalists betting heavily on disruption in the fintech sector. This shows that artificial intelligence is no longer confined to the big tech universe. It’s penetrating financial, healthcare, agricultural, and industrial sectors at a pace that very few predicted two or three years ago.
And the big question hanging in the air isn’t whether artificial intelligence will keep advancing — that answer is already known. The real question is whether people, companies, and institutions can keep up with this pace with the maturity the moment demands. That’s the true test of the adaptation happening right now, in real time, for everyone. 💡
