Amazon's AI-Powered Alexa Faces Hurdles in Solving Hallucination Problem

Amazon Says All It Needs to Do Before Releasing an AI-Powered Alexa Is to Solve the Giant Engineering Problem That Nobody Else on Earth Has Been Able to Solve

Amazon is still hard at work in its efforts to realize an AI-powered Alexa digital assistant. As the Financial Times reports, the tech giant still has to sort out "several technical hurdles" before rolling out the long-awaited feature. One of them, according to Amazon AI team lead Rohit Prasad, is solving the pesky issue of "hallucinations" -- an industry term denoting the non-factual hallucinations that large language models often spit out, to the chagrin of many companies attempting to commercialize the tech. "Hallucinations have to be close to zero," Prasad told the FT. The issue? That's far easier said than done. Despite billions of dollars of investment and the construction of massive data centers to power increasingly complex AI models, even the most advanced chatbots still have a strong tendency to "hallucinate" false claims. Some experts have long argued that the issue might be intrinsic to the tech itself. In other words, hallucinations may always be a part of the equation -- an unfortunate reality that tech companies are unlikely to admit, especially in the face of all of generative AI's buzz right now. "They're really just sort of designed to predict the next word," Anthropic cofounder and president Daniela Amodei told the Associated Press back in 2023. "And so there will be some rate at which the model does that inaccurately." Meanwhile, tech companies that have invested astronomical sums in the tech's development are attempting to sweep the topic of hallucinations under the rug, insisting that it's only a matter of time until the issue is solved. Some companies, like Microsoft, believe that more AI could be the answer. The company unveiled a tool last year that uses AI to evaluate the outputs of other models, though experts have warned that the strategy may be inherently flawed. "Trying to eliminate hallucinations from generative AI is like trying to eliminate hydrogen from water," University of Washington PhD candidate Os Keyes told TechCrunch last year. "It's an essential component of how the technology works." Amazon is several years behind the competition when it comes to releasing a generative AI-powered personal assistant. That's despite having worked on an Alexa redesign since late 2022, according to the FT. The long delay highlights just how difficult it is to overcome the issue of hallucinations -- if it's indeed possible. So far, Alexa's capabilities remain severely limited compared to AI chatbots like ChatGPT, only helping users with simple tasks like changing the music or starting timers. Making matters even more difficult is the fact that keeping AI assistants running while getting "billions of requests a week," according to Prasad, can become extremely expensive due to the tech's power-hungry and highly energy-inefficient nature. That means turning the assistant into a money maker, even with a rumored $10 monthly subscription fee, could prove difficult. And it's not just Amazon. Apple's Siri home assistant is also set to get an AI-powered revamp. But the tech giant is giving itself until 2026 to release it, and in the meantime, Apple had to halt an AI news summary feature this week after it consistently spread hallucinated fake news to millions of iPhone users for over a month. The stakes are high for the likes of Amazon and Apple: unleashing a hallucinating assistant inside your home and giving it access to data from other internet-connected devices, such as smart doorbells or security cameras, could prove disastrous.

Amazon must solve hallucination problem before launching AI-enabled Alexa

Amazon is gearing up to relaunch its Alexa voice-powered digital assistant as an artificial intelligence "agent" that can complete practical tasks, as the tech group races to resolve the challenges that have dogged the system's AI overhaul. The $2.4 trillion company has for the past two years sought to redesign Alexa, its conversational system embedded within 500 million consumer devices worldwide, so the software's "brain" is transplanted with generative AI. Rohit Prasad, who leads the artificial general intelligence (AGI) team at Amazon, told the Financial Times the voice assistant still needed to surmount several technical hurdles before the rollout. This includes solving the problem of "hallucinations" or fabricated answers, its response speed or "latency," and reliability. "Hallucinations have to be close to zero," said Prasad. "It's still an open problem in the industry, but we are working extremely hard on it." The vision of Amazon's leaders is to transform Alexa, which is currently still used for a narrow set of simple tasks such as playing music and setting alarms, to an "agentic" product that acts as a personalized concierge. This could include anything from suggesting restaurants to configuring the lights in the bedroom based on a person's sleep cycles. Alexa's redesign has been in training since the launch of OpenAI's ChatGPT, backed by Microsoft, in late 2022. While Microsoft, Google, Meta, and others have quickly embedded generative AI into their computing platforms and enhanced their software services, critics have questioned whether Amazon can resolve its technical and organizational struggles in time to compete with its rivals. According to multiple staffers who have worked on Amazon's voice assistant teams in recent years, its effort has been beset with complications and follows years of AI research and development. Several former workers said the long wait for a rollout was largely due to the unexpected difficulties involved in switching and combining the simpler, predefined algorithms Alexa was built on, with more powerful but unpredictable large language models.

Amazon Hits Snags in Revamping Alexa Into Agentic AI | PYMNTS.com

Two years into its planned revamp of Alexa into a smarter voice assistant with generative artificial intelligence (AI) at its core, Amazon is running into deployment roadblocks that range from hallucinations to organizational challenges. In an interview with the Financial Times, Rohit Prasad, who leads the artificial general intelligence (AGI) team at Amazon, said the company is still trying to solve problems such as hallucinations, where the AI model makes things up, as well as latency, reliability and other issues. Hallucinations have to be "close to zero," Prasad said. Since large language models are probabilistic, they can hallucinate when asked questions or encounter scenarios outside their training data. That means when a customer orders from Alexa, the agentic AI assistant might purchase another product or invent the quantity of orders, for example. A stickler is that the eCommerce giant has to tread carefully because of Alexa's global scale: It is used in half-a-billion devices in real time. Prasad said this was an "unprecedented" scale for AI assistants. Today, the goal is not only to enable a GenAI-powered Alexa but also make it agentic so it can accomplish tasks in addition to providing information, according to the FT. To be sure, Amazon's competitors have rolled out their own GenAI-powered virtual assistants to many users globally, although these are not yet agentic AI. Prasad said one complexity in making Alexa into an AI agent that operates at scale in real time is that it has to be able to call hundreds of third-party software and services to complete its tasks. "Sometimes we underestimate how many services are integrated into Alexa, and it's a massive number. These applications get billions of requests a week, so when you're trying to make reliable actions happen at speed  ... you have to be able to do it in a very cost-effective way," Prasad said. While costs to run GenAI have been coming down, it can still be expensive to run at scale. Pre-training the foundation models can be costly, but inference -- when the AI model applies its training on new data -- doesn't come cheap either. Amazon thought of charging a subscription for an LLM-powered Alexa or taking a cut of eCommerce sales, a former employee told the FT. But it is the technical hurdles that are holding back most of the progress. Making Alexa smarter is not as simple as adding a large language model to it to replace its simpler algorithms. (Amazon recently introduced its own family of Nova foundation models, but did not disclose their parameter sizes.) "It's not as simple as moving from one model to another," Mike Finley, CTO and co-founder of AnswerRocket, told PYMNTS. "Agentic AI is a bit more nuanced. It needs more structure and guidance to get us a better result. We will have to give it the original 'prompt' like we would in the past, but there's more work to shape the AI behavior we want." Moreover, "for Alexa to level-up in usefulness, we're going to have to trust it more. Would you trust Alexa to send the babysitter some cash? What if it hallucinates a couple of zeros on the wrong side? Agentic models can use resources like the ability to access databases or financial accounts, websites, documents, spreadsheets. But powering Alexa with those tools means gaining consumer trust to click 'allow,'" Finley said. Alexa falling behind its tech giant competitors has irked former Alexa research scientist Mihail Eric, who posted on X in June that Amazon had dropped the ball. He said Alexa had been ahead of the pack, but frittered away its lead due to disorganization at the company, thinning engineering teams and competitive infighting among decentralized teams. Amazon's practice of protecting customer data with guardrails, which Eric acknowledged is a "crucial" policy, also nevertheless meant that the internal infrastructure for developers was "agonizingly painful to work with." It would take weeks to access any data for analysis or experimentation, he added. Eric recommended the following ways to accelerate Alexa's development: invest in robust developer infrastructure, especially around access to compete, data quality assurance and streamlined data collection processes; make LLMs the fundamental building block of the dialogue flows; and ensure product timelines don't dictate science research time frames. An Amazon spokesperson told PYMNTS: "Our vision for Alexa is to build the world's best personal assistant. Generative AI offers a huge opportunity to make Alexa even better for our customers, and we are working hard to enable even more proactive and capable assistance on the over half-a-billion Alexa-enabled devices already in homes around the world."

Amazon's AI lead says technical issues are holding back Alexa AI

Amazon had been planning to roll out a new Alexa powered by generative AI in October 2024, but that obviously didn't happen. According to reports that came out back then, the company pushed back its new voice assistant's release to sometime this year. Now, a new report by The Financial Times says the company still needs to be able to overcome "several technical hurdles" before it can launch a more powerful version of Alexa. One of the main problems it has to solve is "hallucinations," which are incorrect or false results that generative AIs produce at times. Hallucinations have to be "close to zero," Rohit Prasad, leader of Amazon's artificial general intelligence (AGI) team told FT. Since people tend to use Alexa throughout the day, it could end up spitting out a lot of false information if Amazon fails to address the issue. Prasad admits that hallucinations are "still an open problem in the industry," but his team is "working extremely hard on it." Amazon also has to work Alexa's response speed or latency, because users expect to get a response quickly after they ask the assistant a question or after they ask it to perform a task. The Amazon AGI lead said that getting Alexa to that last mile has been really hard. "Sometimes we underestimate how many services are integrated into Alexa, and it's a massive number," he told FT. His team has to ensure that the new assistant will be able to work with hundreds of third-party apps and services. The new Alexa is expected to be powered by Anthropic's Claude AI and the company's in-house Amazon Nova models, and it will reportedly require a subscription as a way for the company to make money. But it still has no solid release date, and based on what a current employee told the publication, it's not rolling out anytime soon. Amazon still has a lot of things to do, they said, such as making sure it works "close to 100 percent of the time," adding child safety filters and testing Alexa various integrations.

Amazon's upcoming Alexa AI brain transplant might make you use it more than just weather and timers

When the new Alexa arrives the hope is it will be used for more than just basic tasks. Amazon has spent years extolling Alexa's abilities as a voice assistant, even though it seems most people use it mainly to set timers and check the weather. Even so, that hasn't stopped Amazon from plotting a far bigger place for Alexa in your life. Amazon wants Alexa to graduate from her relatively simple life of timers and trivia into the AI big leagues as a true personal concierge by leveraging the latest AI models, as Amazon's artificial general intelligence (AGI) leader Rohit Prasad explained to the FT. Prasad and Amazon want to fully transform Alexa's brain through a kind of 'transplant' to swap out the old question-answering engine for generative AI models. If all goes according to Amazon's ambitious plan, Alexa 2.0 will be the digital butler constantly promised, rather than an audio stopwatch and remote control. Prasad admitted it won't be easy, though he is confident Amazon can defeat the obstacles in the way. If he isn't having hallucinations, Alexa needs to eliminate any hallucinations of its own created by the AI. An assistant that fabricates responses that sound plausible but are completely wrong isn't going to get a lot of use. When you're asking about the best route to the airport, "plausible but wrong" isn't going to cut it. Further, Alexa needs to be reliable if people ask it to do more than just play their favorite music. The wrong song is no big deal, but if you request it to book a table for dinner, adjust your lights, and double-check your babysitter's arrival time, you need to be confident it won't get anything wrong. At the same time, caution about mirages can't slow down responses. According to Prasad, while Alexa responds pretty quickly now, the new AI brain is a bit slower, sometimes taking up to ten seconds to answer a query. The company will need to bring the new Alexa up to speed to make it attractive to users. One thing Amazon is particularly keen on is keeping Alexa's personality intact. Prasad said Amazon is hiring experts to fine-tune her voice, diction, and overall personality to make the transition to a more conversational AI. That said, generative AI is probabilistic, meaning it predicts responses based on patterns rather than absolute truths. That makes it great for casual conversations but a bit dicey for high-stakes tasks like managing smart homes or relaying emergency alerts. The stakes are high, and any misstep could hurt Alexa's reputation. No matter how good the new Alexa is at helping users, there's a very obvious issue facing Amazon's plans to make Alexa the ultimate digital concierge. Microsoft, OpenAI, Google, Meta, and others are working toward many of the same goals. In particular, Google has all but overwritten Google Assistant with Gemini across the board. Amazon had a major lead over its rivals regarding smart speakers and smart displays. However, that may not matter if no one thinks to use Alexa when they can turn to Gemini, ChatGPT, or other assistants with similar abilities. Still, Amazon has some assets that could make up any existing gap. The company recently debuted the Nova AI models, which were built in-house and designed specifically for Alexa. Amazon has also deepened its partnership with Claude AI developer Anthropic, bolstered by $8 billion in investment funds. Whether this is enough to leapfrog the competition remains to be seen, but time will certainly tell.

Amazon races to transplant Alexa's 'brain' with generative AI

Amazon is gearing up to relaunch its Alexa voice-powered digital assistant as an artificial intelligence "agent" that can complete practical tasks, as the tech group races to resolve the challenges that have dogged the system's AI overhaul. The $2.4tn company has for the past two years sought to redesign Alexa, its conversational system embedded within 500mn consumer devices worldwide, so the software's "brain" is transplanted with generative AI. Rohit Prasad, who leads the artificial general intelligence (AGI) team at Amazon, told the Financial Times the voice assistant still needed to surmount several technical hurdles before the rollout. This includes solving the problem of "hallucinations" or fabricated answers, its response speed or "latency", and reliability. "Hallucinations have to be close to zero," said Prasad. "It's still an open problem in the industry, but we are working extremely hard on it." The vision of Amazon's leaders is to transform Alexa, which is currently still used for a narrow set of simple tasks such as playing music and setting alarms, to an "agentic" product that acts as a personalised concierge. This could include anything from suggesting restaurants to configuring the lights in the bedroom based on a person's sleep cycles. Alexa's redesign has been in train since the launch of OpenAI's ChatGPT, backed by Microsoft, in late 2022. While Microsoft, Google, Meta and others have quickly embedded generative AI into their computing platforms and enhanced their software services, critics have questioned whether Amazon can resolve its technical and organisational struggles in time to compete with its rivals. According to multiple staffers who have worked on Amazon's voice assistant teams in recent years, its effort has been beset with complications and follows years of AI research and development. Several former workers said the long wait for a rollout was largely due to the unexpected difficulties involved in switching and combining the simpler, predefined algorithms Alexa was built on, with more powerful but unpredictable large language models. In response, Amazon said it was "working hard to enable even more proactive and capable assistance" of its voice assistant. It added that a technical implementation of this scale, into a live service and suite of devices used by customers around the world, was unprecedented, and not as simple as overlaying a LLM on to the Alexa service. Prasad, the former chief architect of Alexa, said last month's release of the company's in-house Amazon Nova models -- led by his AGI team -- was in part motivated by the specific needs for optimum speed, cost and reliability, in order to help AI applications such as Alexa "get to that last mile, which is really hard". To operate as an agent, Alexa's "brain" has to be able to call hundreds of third-party software and services, Prasad said. "Sometimes we underestimate how many services are integrated into Alexa, and it's a massive number. These applications get billions of requests a week, so when you're trying to make reliable actions happen at speed . . . you have to be able to do it in a very cost-effective way," he added. The complexity comes from Alexa users expecting quick responses as well as extremely high levels of accuracy. Such qualities are at odds with the inherent probabilistic nature of today's generative AI, a statistical software that predicts words based on speech and language patterns. Some former staff also point to struggles to preserve the assistant's original attributes, including its consistency and functionality, while imbuing it with new generative features such as creativity and free-flowing dialogue. Because of the more personalised, chatty nature of LLMs, the company also plans to hire experts to shape the AI's personality, voice and diction so it remains familiar to Alexa users, according to one person familiar with the matter. One former senior member of the Alexa team said while LLMs were very sophisticated, they come with risks, such as producing answers that are "completely invented some of the time". "At the scale that Amazon operates, that could happen large numbers of times per day," they said, damaging its brand and reputation. In June, Mihail Eric, a former machine learning scientist at Alexa and founding member of its "conversational modelling team", said publicly that Amazon had "dropped the ball" on becoming "the unequivocal market leader in conversational AI" with Alexa. Eric said despite having strong scientific talent and "huge" financial resources, the company had been "riddled with technical and bureaucratic problems", suggesting "data was poorly annotated" and "documentation was either non-existent or stale". According to two former employees working on Alexa-related AI, the historic technology underpinning the voice assistant had been inflexible and difficult to change quickly, weighed down by a clunky and disorganised code base and an engineering team "spread too thin". The original Alexa software, built on top of technology acquired from British start-up Evi in 2012, was a question-answering machine that worked by searching within a defined universe of facts to find the right response, such as the day's weather or a specific song in your music library. The new Alexa uses a bouquet of different AI models to recognise and translate voice queries and generate responses, as well as to identify policy violations, such as picking up inappropriate responses and hallucinations. Building software to translate between the legacy systems and the new AI models has been a major obstacle in the Alexa-LLM integration. The models include Amazon's own in-house software, including the latest Nova models, as well as Claude, the AI model from start-up Anthropic, in which Amazon has invested $8bn over the course of the past 18 months. "[T]he most challenging thing about AI agents is making sure they're safe, reliable and predictable," Anthropic's chief executive Dario Amodei told the FT last year. Agent-like AI software needs to get to the point "where . . . people can actually have trust in the system", he added. "Once we get to that point, then we'll release these systems." One current employee said more steps were still needed, such as overlaying child safety filters and testing custom integrations with Alexa such as smart lights and the Ring doorbell. "The reliability is the issue -- getting it to be working close to 100 per cent of the time," the employee added. "That's why you see us . . . or Apple or Google shipping slowly and incrementally." Numerous third parties developing "skills" or features for Alexa said they were unsure when the new generative AI-enabled device would be rolled out and how to create new functions for it. "We're waiting for the details and understanding," said Thomas Lindgren, co-founder of Swedish content developer Wanderword. "When we started working with them they were a lot more open . . . then with time, they've changed." Another partner said after an initial period of "pressure" that was put on developers by Amazon to start getting ready for the next generation of Alexa, things had gone quiet. An enduring challenge for Amazon's Alexa team -- which was hit by major lay-offs in 2023 -- is how to make money. Figuring out how to make the assistants "cheap enough to run at scale" will be a major task, said Jared Roesch, co-founder of generative AI group OctoAI. Options being discussed include creating a new Alexa subscription service, or to take a cut of sales of goods and services, said a former Alexa employee. Prasad said Amazon's goal was to create a variety of AI models that could act as the "building blocks" for a variety of applications beyond Alexa. "What we are always grounded on is customers and practical AI, we are not doing science for the sake of science," Prasad said. "We are doing this . . . to deliver customer value and impact, which in this era of generative AI is becoming more important than ever because customers want to see a return on investment."

Report: Amazon to Refashion Alexa Into AI Agent | PYMNTS.com

In the new role, Alexa would be able to complete practical tasks, although the tech giant must resolve some issues that have plagued the voice-activated system's upgrade, the Financial Times reported Tuesday (Jan. 14). For the last two years, Amazon has worked to redesign Alexa so that its "brain" is replaced with generative AI, the report said. Alexa needs to clear several technical hurdles before launch, such as "hallucinations" or fabricated answers, its response speed and reliability, said Rohit Prasad, head of Amazon's artificial general intelligence (AGI) team, per the report. "Hallucinations have to be close to zero," Prasad said, according to the report. "It's still an open problem in the industry, but we are working extremely hard on it." While Alexa performs a range of functions, such as playing music, setting alarms or answering questions, the company's vision is to turn it into a personalized concierge, the report said. Alexa would suggest restaurants or set lights in a bedroom according to a user's sleep cycles. Multiple workers from Amazon's voice assistant team said the project was saddled with complications after years of AI research and development, according to the report. Former workers said the delayed rollout was chiefly due to the unexpected difficulties of switching and combining the simpler, predefined algorithms Alexa was built on with more powerful but less predictable large language models. Amazon said it is "working hard to enable even more proactive and capable assistance" of its voice assistant, per the report. The PYMNTS Intelligence report "How Consumers Want to Live in the Voice Economy" found that roughly a quarter of American consumers said they would be willing to pay $10 per month to access a virtual assistant that could handle everyday tasks. Meanwhile, in retail applications, Keith Kirkpatrick, research director of enterprise applications at The Futurum Group, told PYMNTS this month: "I think in time, AI agents will impact certain job functions in retail, but most of these tasks are digital, as opposed to physical tasks. As such, many of these functions are still marketing- or commerce-based, and retailers will be looking to companies in adjacent industries to see how well AI agents ... are delivering value to their customers."

Amazon to Rebrand Alexa as an AI Agent

Global e-commerce giant Amazon is set to 'relaunch' its digital assistant Alexa as an artificial intelligence agent. For the past two years, the company has been reportedly working on redesigning Alexa and integrating generative AI. Amazon wanted Alexa to go beyond performing simple tasks like setting alarms and reminders or playing music to more practical tasks using AI. Rohit Prasad, the head scientist at Amazon's AGI (artificial general intelligence) team, said that in order to function as an AI agent, Alexa must be able to call hundreds of third-party software and services. He added that the company's Nova models were built with Alexa's needs in mind. However, several employees revealed that Amazon encountered numerous challenges, involving hallucinations, latency, and reliability, to transform Alexa into an AI agent. "Hallucinations have to be close to zero," said Prasad, adding that the company is 'working extremely hard on it'. Rebranding Alexa as an AI agent is expected to improve the product's functionality and sales. A study from Mountain Research suggests that over 70% of consumers would prefer products that explicitly disclose the use of AI. "While most consumers haven't used AI firsthand, they remain interested or excited in the technology as long as it's not hurting people and brands are being transparent with its usage. The opportunities it offers are virtually limitless," read the report. While Alexa does use AI to accomplish the tasks it can today, there's very little disclosure around this. Voice-based AI agents will surely be on the rise in 2025. At the end of last year, Hume AI, the voice-to-voice 'conversational' assistant, explored a new capability powered by Anthropic Claude's Computer Use. Hume AI converts voice commands to text that Claude can recognise, which is then translated into actions that Computer Use performs. To demonstrate its capabilities, Hume AI was taken to X, where a user verbally instructed Hume AI and Computer Use to play chess on a web browser. Once Alexa is rebranded as an AI agent, it would be interesting to see if Amazon explores Computer Use capabilities. Anthropic was one of the first companies to build an autonomous computer use agent, followed by Microsoft's Copilot Vision and Google's Project Manager. Claude's tool, however, has found the first-mover advantage. Many companies integrate it into their workflows, and developers build unique use cases with it. Amazon is one of Anthropic's largest investors, investing $4 billion in the AI startup in March 2024.

Amazon's Alexa AI Redesign Could Resolve Trust Issues but Must Overcome Technical Challenges

Integrating conversational AI could help increase trust in the platform, but it could backfire if not pulled off correctly. For the last two years, Amazon has been working to upgrade its Alexa voice assistant with generative AI, which is expected to be the biggest overhaul of the software since it was launched more than a decade ago. According to recent reports, the long-awaited upgrade may finally be imminent. But Amazon's challenge is how they integrate new AI functionalities without changing established functions that users are familiar with and, crucially, without losing their trust. Trust and AI: A Complex Equation For the makers of voice assistants like Alexa, integrating modern foundation models can be a double-edged sword. On the one hand, research has shown that users value more human-like interactions. This suggests that new, more conversational AI may be perceived as more trustworthy than the previous generation of formulaic voice response systems. On the other, the problem of AI hallucinations limits how much people can trust large language model (LLM) outputs. In comments reported by the Financial Times, one former Amazon employee working on Alexa cautioned that "at the scale that Amazon operates," hallucinations could occur "large numbers of times per day," damaging the firm's brand and reputation. Maintaining Reliability As well as damaging people's trust in AI systems, hallucinations also reduce reliability. Today, Alexa integrates hundreds of applications and can reliably carry out countless different actions. If the platform's new LLM "brain" fails to live up to the standard of reliability users have come to expect from Alexa, it would undermine the entire upgrade. Due to the sheer number of processes Alexa manages, overhauling the entire system is a mammoth undertaking. Amazon's AGI Lead Rohit Prasad told the Financial Times that the firm still had to overcome several technical hurdles before rolling out the AI upgrade. As he noted, the "last mile" of development is inevitably "really hard" as the firm looks to launch the integration at scale. To achieve the right level of performance efficiently, Amazon will reportedly deploy a suite of AI models, including Anthropic's Claude and its own Nova models under the hood. However, the company still hasn't committed to a release date. The Risk of Rushing AI While Alexa's AI makeover has been more than two years in the making, rushing it out before it is ready risks making a product used by millions of people worse. Consider for example, the case of Google's smartphone assistant. When Google switched the default Android assistant from Google Assistant to the AI-powered Gemini last year, the move was met with frustration by users because it failed many of the basic tasks its predecessor could perform. With their ability to understand natural language, LLM-based assistants remove the need for specific commands, enabling more naturalistic interactions. But if they can't perform the same tasks as the previous generation of command-based virtual assistants, their enhanced conversational abilities are all for nothing.