Last Week in AI: DOGE uses AI for gov’t; Amazon launches Alexa

How DOGE is using AI to downsize the government

Elon Musk’s DOGE is allegedly using artificial intelligence (AI) in its mission to reduce the size of the federal government. According to recent reports, Musk—who recently sent an email to federal employees asking them to list five things they accomplished that week—plans to feed those responses into a large language model (LLM) to determine whether their work is mission critical.

Unsurprisingly, DOGE’s actions created controversy, and critics accused Musk and the DOGE team of using their email initiative to justify mass layoffs in the federal government. Musk, in response, clarified that this first email was more of a litmus test. He said any response, no matter how trivial, was enough to confirm that a living, breathing person was behind the keyboard. He emphasized that the real purpose of the email was to determine whether the individual receiving a government salary was competent enough to respond. According to Musk, the real red flag is if an employee fails to respond altogether.

U.S. President Donald Trump was quick to voice his support for Musk’s initiative, calling it “great” and emphasizing the need for outside intervention in what he described as a bloated and inefficient government.

While AI can be a tool in DOGE’s mission, it’s questionable whether it should be the only tool used to determine whether someone’s role is essential. For an LLM to make truly informed decisions, it would need to be trained on private government data—details about departmental missions, operational processes, and key objectives that are unlikely to be publicly available. Without that context, it’s unclear how effective DOGE’s AI-driven assessment will actually be.

Is the demand for AI decreasing?

Is AI demand cooling off? This week, Microsoft’s (NASDAQ: MSFT) decision to cancel several data center leases and scale back by a few hundred megawatts of capacity in data centers caused analysts to question whether the demand for artificial intelligence infrastructure is slowing down or if it was overblown, to begin with.

One possible explanation for Microsoft decreasing the bandwidth it needs from data centers could be OpenAI’s recent partnership with Oracle, announced in the summer of 2024. As a result of this partnership, OpenAI will be relying more on Oracle Cloud Infrastructure instead of Microsoft’s infrastructure, which will likely reduce Microsoft’s anticipated AI workload, making the additional capacity unnecessary.

Amazon launches AI-infused Alexa—but will consumers care?

On February 26, Amazon (NASDAQ: AMZN) announced Alexa+, its latest AI-enhanced virtual assistant version. Described as more conversational, capable, and smarter than ever, Alexa+ integrates large language models (LLMs), allowing it to summarize documents, organize schedules, answer complex questions, and even create quizzes from uploaded materials. It also features agentic capabilities, which Amazon calls “experts,” allowing Alexa+ to browse the web and complete complex tasks autonomously from end to end—such as booking restaurant reservations.

Although Amazon probably sees its new product launch as a win for the company, I think the launch of Alexa+ raises a few questions, especially given the product’s history.

For starters, Amazon’s devices business—which includes Alexa—has lost the company over $25 billion in recent years. Surprisingly, Alexa+ is being offered as a free product to Amazon Prime subscribers and for $19.99 to non-members, which is a strange move amid the department’s losses. This is especially puzzling given that reports previously suggested Amazon was working on a new Alexa version designed to be a revenue driver—ultimately revealed as Alexa+. This makes it an odd choice to offer it for free to Prime users.

Beyond that, I think an even bigger question is whether consumers even want AI-powered devices. Time and time again, the market has indicated that most AI-infused consumer gadgets fail to gain traction. Smart assistants like Alexa and Google Assistant have struggled to deliver meaningful value beyond basic voice commands. AI-integrated hardware often lacks compelling features that can’t already be found in existing mobile or desktop AI tools.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: Demonstrating the potential of blockchain’s fusion with AI