Tech Brew keeps business leaders up-to-date on the latest innovations, automation advances, policy shifts, and more, so they can make informed decisions about tech.

Tech Brew

Patrick Kulp

techbrew.com

Accenture’s chief AI officer thinks every C-suite needs a job like hers

Lan Guan is leading the consultancy’s efforts to harness a huge wave of AI interest from the business world.

A video of a sailboat moving down a river generated with Runway's Gen-2

Video generated with Runway's Gen-2 of a "person running on a track."

Video generated from the prompt "cat running on the street."

Is video the next frontier for generative AI?

AI startups want to make inroads in Hollywood, but the technology still has its limits.

Persistent AI hype fails to buoy startup market in Q1

PitchBook data shows a sluggish market for venture capital investing and fundraising.

Every time you pop an idle thought into ChatGPT or another AI interface, it triggers a far-flung journey. Your query—“What is this rash?” perhaps—races to a hulking data center in a place like Abilene, Texas, where it’s routed to the right stack of servers.

The traffic directors of all this activity are a series of high-bandwidth network switches—devices that coordinate data between servers—made by companies like Broadcom. The AI wave has reshaped Broadcom’s semiconductor business; its switches are now designed to support the high-capacity needs of massive AI data centers, and its custom AI accelerator line is a fast-growing source of revenue.

“Over the last three years, things have changed pretty dramatically with AI,” Peter Del Vecchio, product line manager for Broadcom’s Tomahawk family of data center switches, told us. “For example, we announced [Tomahawk 5] in August of 2022, and the main market at that time was just general data center networking. ChatGPT was announced three months later, and pretty much everything changed.”

 with OpenAI to jointly build 10 gigawatts’ worth of AI accelerators over the coming years, tapping both Broadcom’s custom AI accelerators and its networking infrastructure. It’s one of 

 of infrastructure partnerships that OpenAI has forged over the past several months.

On the heels of this new deal, Tech Brew took a behind-the-scenes tour of Broadcom’s network switch R&D labs, where engineers undertake the multiyear process of assembling these devices.

The labs are situated in the ground-floor wing of a nondescript office building in Broadcom’s San Jose, California, campus. The rooms are crowded with workbenches and scattered with specialized devices. Thick bundles of wires and cooling pipes thread through ceiling vents to drape down to each workstation.

“This is basically our factory, where we’re building these things and testing them and characterizing them, all in that room,” David Hollis, principal mechanical and thermal engineer at Broadcom, said. “We’ll have technicians, engineers, program managers all coalesce around this hub and monitor where we’re at with all the build…Essentially, it’s a one-stop shop for all these systems to build, assemble, debug, and deploy.”

Around 1,400 employees work on Broadcom’s switch team, about half of them focused on the software side, according to Del Vecchio.

Switches take around three years to develop from initial design to release, so engineers at the beginning of the process don’t necessarily know what kinds of demands the AI world will eventually place on them, Del Vecchio said.

“We try to build quite a bit of flexibility in there, and we can make some tweaks along the way,” he said. “[But] the basic architecture is going to be pretty much locked in a year and a half, two years before we actually sample to customers.”

In one lab, a team uses machines to swiftly raise or lower the temperature via attached hose-like appendages. Some of Broadcom’s chips advertise extreme durability; they’re designed to withstand the subzero chill or scorching heat of somewhere like a remote cell tower. That room is where engineers make sure those claims hold up.

“These devices can reject up to about 1,500 watts of power over a very small area,” Hollis said. “This system is super-cooling a non-conductive fluid…You don’t want to get this fluid on your hand.”

A temperature forcing machine in Broadcom's lab

Temperature is a major concern across the whole lab, especially right now. Broadcom is currently undertaking an upgrade involving a liquid cooling system that will pipe water and coolant around hoses overhead in a closed loop. The space is a bit messy as a result; engineers keep telling us that it’s usually much tidier.

“We’re still going through some of the transition,” Greg Barsky, who manages hardware R&D for Broadcom’s Core Switching Group, shouts over the roar of hundreds of fans.

We’re standing amid columns of stacked servers festooned with ethernet cables and translucent cable hoses at the ready for said transition. For now, ear protection is required to enter the small data center’s jet-engine din, loosely contained by a border of plastic flaps. The data center powers development and quality assurance for the software available with the switches.

The switch to liquid cooling will ultimately make the lab more power-efficient than the current fans. “Liquid is always going to be better to move heat, and it’s going to enable, eventually, smaller data centers and less power consumption because, well, fans [use] a lot of power,” Hollis said.

Data centers are increasingly making this switch, as air cooling is no longer able to handle the heat created by higher-powered AI workloads.

Much of what has changed about Broadcom’s switch business over the last few years is a result of the unprecedented scale of new AI data centers, Del Vecchio said. The constant increase in compute needs creates new challenges for networking between models sprawled across different parts of data centers or even multiple distant buildings.

“These models are so large and there’s so much training that goes on with them that [they’re] having problems finding power from the power companies, so what they’ll do is say, ‘OK, I’ll get this data center over here, and 60 miles away, I’ll put in another data center,” Del Vecchio said. “Whereas before, it was that you [would] have a model that would run on one GPU…Now, people are saying, ‘OK, now I’ve got to scale up or scale out to maybe on the order of 32,000 to 64,000 GPUs within a building. And now I need to connect multiple buildings together in order to be able to do this training. The scale of these things, the amount of power is just insane.”

Keep your eyes peeled for part two of this feature in the coming days for an in-depth discussion of the big bets Broadcom is making with its custom AI chip business.

A small data center in Broadcom's switch design labs.

A behind-the-scenes look at Broadcom’s design labs

We toured the San Jose campus where Broadcom builds network devices to run AI data centers.

A story played out in Wednesday’s Big Tech earnings bonanza that’s 

Microsoft, Alphabet, and Meta all surpassed topline analyst expectations around earnings and revenue. All three also upped their estimates of the massive amounts they had spent or plan to spend on AI infrastructure.

Investors were left to pick a narrative. Wall Street cheered Alphabet’s results with an upward swing in share price, while Meta’s share price plummeted (the company also incurred a big 

 that cratered its profit), and Microsoft shares dipped.

Microsoft clocked $78 billion in revenue and $28 billion in profit; Google reported $102 billion, with profits of $35 billion; and Meta saw $51 billion in revenue and $2.7 billion in profit (after that $16 billion tax expense).

 of late, and that loomed over all three earnings calls. But the actual B word only cropped up once, in an analyst question to Microsoft execs: “How much confidence do you have that the software, even the consumer internet business, can monetize all the investments we’re seeing globally, or frankly, are we in a bubble?”

Microsoft CFO Amy Hood said persistent strong demand justifies the spending, while CEO Satya Nadella said the data center investments are flexible enough to adapt to different kinds of workloads.

“I look at the entirety of these high-value agent systems, and when we look at the efficiency of and fungibility of our fleet, that’s what gives us the confidence to invest both the capital and the R&D talent to go after this opportunity,” Nadella said during the call.

As far as infrastructure spending, Hood said “total spend will increase sequentially” and spending will accelerate next year.

Meta has been busily remaking its AI research arm through 

. CEO Mark Zuckerberg faced questions about monetizing his company’s quest for AGI.

Zuckerberg claimed Meta AI has seen engagement grow, including a tenfold increase in media generation since the launch of AI social video app Vibes in September. Advertising dollars will eventually follow that engagement, he said. He claimed the advertising process will eventually be fully automated.

“Advertisers are increasingly just going to be able to give us a business objective and give us a credit card or bank account, and have the AI system basically figure out everything else that’s necessary, including generating video or different types of creative that might resonate with different people that are personalized in different ways,” Zuckerberg said.

That will come at a heavy price. Meta increased its spending outlook slightly for the year to $116 billion–118 billion (up from $114 billion–$118 billion). Expect that number to be “notably larger” in 2026, CFO Susan Li said.

 Demand for Google Cloud services drove strong growth, with a 34% YoY jump to $15 billion in revenue. Google CFO Anat Ashkenazi said that led the company to also adjust its spending upward to $91 billion–$93 billion from a previous estimate of $85 billion.

Google Chief Business Officer Philipp Schindler said new AI tools were continuing to transform the search experience.

“AI mode continued to drive growth in overall queries, including commercial queries, really creating more opportunities for monetization,” Schindler said. “AI overviews are scaling up and are working for our entire user base.”

Google, Meta, Microsoft on 3 pillars surrounded by binary code

Big Tech doubles down on AI spending amid bubble fears

Meta, Microsoft, and Google all beat revenue expectations and upped data center spend.

When it comes to aggregating the news, ChatGPT and its ilk could use some copy editors.

A far-reaching study from the BBC and the European Broadcasting Union found that current affairs-related summaries from four major AI chatbots were rife with errors. Nearly half (45%) of more than 3,000 responses contained at least one significant issue, including sourcing problems (31%), factual inaccuracy (20%), or a lack of context (14%).

The research spanned 22 public service media organizations across 18 countries and 14 languages. Journalists at each of these orgs prompted ChatGPT, Perplexity, Google Gemini, and Microsoft Copilot for news, then analyzed the responses.

The authors used the findings to create a comprehensive toolkit aimed to boost AI news integrity for tech companies, media orgs, researchers, and the public.

 for informational needs, the news outlets wanted to understand what that means for the media ecosystem, as well as their own reputational risks if content is being distorted.

“This research conclusively shows that these failings are not isolated incidents,” EBU Media Director and Deputy Director General Jean Philip De Tender said in a statement. “They are systemic, cross-border, and multilingual, and we believe this endangers public trust. When people don’t know what to trust, they end up trusting nothing at all, and that can deter democratic participation.”

The new report builds on a similar but smaller 

 by the BBC in February, which found that 51% of AI answers to questions about news contained errors.

Accuracy levels varied across different chatbots. Gemini performed by far the worst, with 76% of responses containing significant issues—double the error rate of the next highest, Copilot, at 37%. ChatGPT clocked in at 36% and Perplexity at 30%. The bulk of Gemini’s failures seemed to be sourcing issues: 72% of responses had some form of misattribution (including a lack of any sources at all), compared to 24% from ChatGPT and 15% from both Perplexity and Copilot.

 Chatbots sometimes struggled to properly cite laws; Perplexity claimed that surrogacy is illegal in the Czech Republic, for instance, when it is in fact unregulated. And information was often outdated; chatbots struggled to keep up with the latest European leaders and the new pope.

AIs had an especially hard time with breaking or complex news stories, even after they were no longer developing.

“The data suggests that assistants particularly struggle with fast-moving stories with rapidly changing or updating information (Trump trade war, Myanmar); intricate timelines involving multiple actors (Yemen); detailed information (China exports, Trump trade war, Trump tariffs); or topics that require clear distinction between facts and opinions and proper attribution of claims (Orbán, climate change),” the authors wrote.

 In addition to creating misinformation and posing brand risks to news outlets, these accuracy problems may actually be hindering the use of AI chatbots for news consumption as well. A 

 earlier this month found that only 2% of Americans regularly get news from chatbots (7% said “sometimes”). Those numbers stand in contrast to the 

 of Americans who use ChatGPT in general (34% as of this June).

Respondents also seem keenly aware of the factual issues at play. A third of Americans who use chatbots for news said they find it difficult to determine what’s accurate, Pew reported, and about half said they come across news they think is inaccurate.

A close-up of an iPhone screen with red incorrect icons interspersed over an AI summary.

News summaries from AI chatbots have major accuracy problems

A study from the BBC and EBU found that 45% of responses had significant issues.

A year ago, Salesforce CEO Marc Benioff announced a huge bet on AI agents amid the cartoonish sprawl of his company’s annual Dreamforce conference, which resembles something of a Disneyland for B2B SaaS enthusiasts.

This year, the sprawling San Francisco convention was remade in the image of Agentforce; the term “agent” was mentioned in around three-quarters of the 1,443 sessions. The company says 12,000 customers are now using the autonomous AI builder platform.

That may sound like a lot, but Salesforce has more than 150,000 customers, meaning 92% have yet to adopt Agentforce.

—as the next big chapter in AI adoption in the office. But one of the biggest obstacles for companies is trust: Handing over business processes to free-acting AI is a big ask, 

Salesforce VP of AI and Agentforce Nancy Xu told Tech Brew that building agents that are predictable at scale is a major focus for her employer. It’s also one of the biggest barriers for companies that may have tried agents as a proof of concept, but encountered snags around reliably integrating business processes when fully deployed, she said.

“We have customers that go to the big LLM companies, like, ‘Oh, this is great. I just can plug in. It’s so easy, I get an agent working off the bat,’” Xu said. “But if it doesn’t verify your user’s identity first and follow your business processes, it’s not an agent you can use at scale…Yes, they can get the initial implementation. The agent can say things that conversationally make sense, but then if they don’t follow the processes, it doesn’t scale up.”

Salesforce’s latest version of its agent program, Agentforce 360, has features designed to improve predictability, like a new programming language called Agent Script that allows for “more determinism and more control,” according to Xu. They’re also supposed to be easier to build so nontechnical subject matter experts can hone them.

When it comes to hallucinations, Salesforce President and Chief Legal Officer Sabastian Niles said during Dreamforce last week that the company thinks about three types of shortcomings: inaccuracy, irrelevance, and a lack of creativity. Sometimes, there can be trade-offs between them, but Agentforce is designed to let customers combine the more creative aspects of generative AI with more deterministic actions, he said.

“If you try to completely eliminate hallucinations, maybe you lose the creativity,” Niles said. “As you bridge those two together—determinism, non-determinism in agentic deployments—you really also should be reducing the set of hallucinations in a way that is effective across accuracy, relevancy, but also bringing forward the right sense of creativity.”

Security concerns are also becoming more complex as companies build networks of multiple agents, according to Brad Arkin, Salesforce’s chief trust officer. Arkin said that as agents interact with multiple other agents and humans in different roles, security concerns multiply.

“How on earth do you capture all of that, and how do we consider unexpected outcomes that may arise from the composition of these very complex capabilities?” Arkin said. “And that composability analysis is something that, as an industry, everyone is anticipating is going to get complicated really fast.”

Salesforce’s Futures team has considered possible bigger-picture scenarios around an agentic future, according to Niles, from “AI trade wars” to “digital sovereignty issues.”

“We’ve talked about, what if we really do end up manifesting the [agent-to-agent] economy,” Niles said. “There’s a lot of different implications about that, and not just on the upside opportunities, but areas that we ourselves take very seriously around trust and ethics.”

A Dreamforce arch at an entrance to the conference

How Salesforce is trying to get companies to trust agents

At Dreamforce, company execs talked about how to make agents more predictable.

 flooding platforms, the book world’s relationship with GenAI has been decidedly rocky.

But OverDrive CEO Steve Potash argues that AI recommendations and other technology can help promote libraries at a time when literacy 

. OverDrive works with more than 90% of libraries in North America and operates the e-book app Libby, the video streaming platform Kanopy, and school library app Sora (no relation to the OpenAI video tool).

OverDrive recently introduced an AI-powered book recommendation feature called Inspire Me that allows readers to find books through prompts or past saved titles. Potash said it’s a way to “explore the nooks and crannies of these fabulous collections our public libraries have invested in.” But it has garnered 

Inspire Me isn’t the end of OverDrive’s AI ambitions; Potash wants it to eventually better match users to content across its platforms.

The founder and CEO of nearly four decades talked to us about how libraries are using technology to find new readers and the threats facing these institutions and reading more broadly.

This conversation has been edited for length and clarity.

With Libby, a lot of people who like books and like reading, they’re maybe more suspicious of AI with these 

. How were you making sure that you were careful about wading into AI? Did you anticipate pushback?

As a certified B Corp, we are accountable to high standards of transparency and business practice, and protecting the planet, and worthy of winning the trust of government funding, and public institutions, like the New York Public Library and Library of Congress.

OverDrive is invested with publishers, with trying to research the future of literacy and overcome reading challenges, and pushing the science of reading to expand the value of access to spoken words or words on the screen. So everything we do is permission-based.

The AI pertaining to this Inspire Me service…is an OverDrive-contained LLM that is only utilizing the cataloging information, metadata, and other controlled data points that we have. And we add value to the catalogs of all of our libraries, or how we enhance the information around every book.

We had a long history of trust, consumer engagement, discovery, and engagement with readers to find their next great read. But with Inspire Me, we’re using all of that 25 years of internal cataloging data that enhances the books.

We should mention it’s purely opt-in. You’re not using it unless you decide you want to play with it. So it is an optional, whimsical way to [search]. I’ll tell you what’s most encouraging to me: Not only are the people that are activating it using it, they’re borrowing books, and we’re analyzing the books that they’re borrowing from the inspirations. And these are books that, before Inspire Me, were not circulating.

 doesn’t need any help. The books that are being circulated from Inspire Me, they’re now unbelievably relevant books for these readers that are available now. And they weren’t coming up as you were browsing, or they weren’t coming up in the higher search results. Or you’ve never heard of this author, or it’s a debut author. So that’s what’s been really exciting. And this is where the libraries are also really pumped.

From the library perspective, when we added into Libby Notify Me and Deep Search [features that allow people to search for and request books that aren’t in a library’s catalog,] it added a whole new dimension for librarians to say, “If we’re going to build the digital circulating collection of our Libby or digital library, we can now ask our readers for what they want.”

So when you think about that, what’s the next step? How do we get folks to discover things that the library already vetted, already purchased for the catalog? It’s in their whole worldwide discovery in the libraries, on their public service in Libby. But it’s not getting used.

And you’re starting to see evidence that is happening? That users are finding lesser known books?

Absolutely. First of all, we’re only showing books that are not on The New York Times best-selling list or being featured on Oprah. We’re only showing books that are available now. All The New York Times bestsellers, they’re all checked out, and there’s a wait list.

After we have the first few million users regularly using [Inspire Me], we’ll be circling back with our library advisors and saying, “How are we doing?”

 at a publishing conference where you talked about how reading has been in decline for decades—

I have data that should scare every author and publisher. It’s a consumer lifestyle product that needs to reverse its decline in relevance. And I’m offering them examples and strategies that I know will work.

You said it’s competing more with short-form video. You’ve also talked about library funding cuts and frustrations with various book censorship actions. How do you stay optimistic amid all this?

What makes me optimistic is…our chief marketing officer I hired and recruited. I said, “I want you to bring our nation’s public libraries 25 million new library card holders. Go get ’em.” She says, “OK, well, we have to grow the audience for the book product. I have to go after non-book users.” Because, let’s face it, the librarians, the authors, the publishers, the booksellers, Goodreads, everybody knows how to sell to that avid adult fiction reader. We know how to reach them. They’re great. But what are they doing to grow the audience for people that don’t use books? And books are used for so many purposes.

So my chief marketing officer said, “We’re going to go after millennial males. We’re going to go after sports fans.” And she started going after [those] demographics. We announced last fall a three-year partnership as the literacy partner to the LA Lakers. So now at the Lakers games, Libby is everywhere. I have access to the Lakers. We’re doing it with the Los Angeles Public Library. They’re taking branches and turning into the Lakers literacy lounge, brought to you by Lakers and Libby…I’m talking to Shaq [O’Neal] and Kareem [Abdul-Jabbar].

How did you identify those as the biggest potential demographics?

We ask our librarians, “Who do you want to reach?” They do demographics; they have heat maps of their service area, broken down by street, by ZIP code. They know who’s coming in the branches, who’s their physical user, who’s their digital user. In every community, [they’re] trying to have relevance, to keep their budgets, to keep their staffing, to keep their branches open. Many of these states are under attack, either [by] the executive orders and the [cuts to the Institute of Museum and Library Sciences] and Congress.

After Covid, when I ask directors, “How can we help?” They say, “We’ve got to get foot traffic back to our central library.”….In some [geographic areas], we are doing very targeted tests. You’re in a community that might have a military base. How do we become relevant to them? Or it’s a college town, and they want to supplement [education]. So it’s asking each individual librarian, “Who is it you’re trying to either bring in, or who is an underserved portion of your community that we might help?”

How are you looking at expanding your use of AI?

AI is helping us integrate all forms of media around things that readers are interested in…One of the things that’s in flight right now is we are integrating all of the information about [Kanopy’s] 30,000-plus feature films, documentary, TV shows, foreign films, [and] children’s read-along storybooks into the catalog that also contains all the world’s greatest literature, books, e-books, audiobooks, comics, manga. Now we’re adding all the video.

So just imagine when you say, “I’m interested in this.”...The future state of AI [in Libby] is going to be, “I’m interested in medieval whatever, the renaissance faire.” I want to then all of a sudden have the library, have Libby, have the library catalog show me everything that I’m looking for. And it could be a film, it could be audio, it could be a graphic novel, a comic, manga, it could be a book. Heck, we’re going to be integrating all forms of the work in print, hard goods, digital, video, streaming.

[Right now,] Inspire Me is looking at just a subset of the library’s collection based around e-books and audiobooks available for you to borrow in Libby. The next step will be, what if it was every book? What if it was every book, every magazine, every newspaper, every video?

A portrait of OverDrive CEO Steve Potash.

Why the Libby e-book app is turning to AI

OverDrive CEO Steve Potash thinks the tech can promote reading amid threats to libraries.

Why this library e-book app is turning to AI

AI is becoming widespread in the workplace, but employees who have no say in how it’s used report lower job satisfaction.

That’s according to one of a handful of recent data points on how AI is shaping the future of work. Any big AI job-pocalypse might be further off than the current panic would indicate, another report found, but that doesn’t mean the technology isn’t reshaping offices in the meantime.

Here’s a roundup of some of the latest research on AI and jobs.

 Plenty of studies tally the number of job openings and employment levels; a new index aims to take a deeper look at the 

The report comes from a group of nonprofits and research orgs—Jobs for the Future, the Families & Workers Fund, the W.E. Upjohn Institute, and Gallup.

It found that 60% of Americans lack quality jobs, with more than half feeling left out of important decisions in the workplace, including the adoption of new technology like artificial intelligence tools, where 55% of respondents said they would like more influence.

This lack of input was consistent across education levels: From “high school or less” to graduate degrees, the portion ranged between 53% and 57%. Those who do have a say in tech adoption tend to be happier—46% of those with desired influence reported satisfaction, versus 28% among those without.

“Employees have opinions about the way that technology is implemented, and they experience the direct impacts of that technology on their jobs,” Molly Blankenship, director of strategy and impact at Jobs for the Future, told Tech Brew.

“The important thing that jumped out to me from conversations with workers in this work is that the conversation around technology right now is not a two-way dialogue, and if we could make it [one], that could have a really major impact, not only on worker satisfaction…but also how successful the actual tech implementation is,” Blankenship added.

 New research from Indeed found that the more exposed a job is to AI replacement, the less likely it is to mention training programs in the job description.

That’s partly because the jobs most exposed to GenAI tend to be higher-skilled and higher-paying, whereas training programs are clustered in jobs with lower wages and lower experience or education requirements, Indeed said. On the whole, jobs advertising training programs have been on the rise since 2018, from 3.4% to 8.1%.

Indeed also released its annual AI at work report last month, which found that more than a quarter of jobs on its platform could be “highly” transformed by AI, while 54% might “moderately” change. The research looked at how technology might alter component skills involved in a role rather than whether AI would replace the job outright.

 When should we expect that transformation? If past technology waves are any indication, it could take longer than some headlines lead you to believe.

 from Yale University’s Budget Lab, which found that despite hand-wringing, the pace of labor market change in the nearly three years since ChatGPT’s release has roughly matched that following other broad innovations, like computers or the internet. Those tech disruptions occurred over decades rather than months or years, the authors wrote, and AI effects are happening on a similar timeline.

“Our metrics indicate that the broader labor market has not experienced a discernible disruption since ChatGPT’s release 33 months ago, undercutting fears that AI automation is currently eroding the demand for cognitive labor across the economy,” the Yale researchers wrote.

An AI robot robot sitting side by side with a businessman at an office desk working

Workers who have less say in AI adoption report lower job satisfaction

We rounded up the most recent data on AI and jobs.

 of companies want to supply AI agents that purport to help you do your job better.

Add Amazon Web Services (AWS) to that growing list. The tech giant is rolling out a new platform called Amazon Quick Suite that aims to provide a universal chat interface for surfacing and assembling workplace information, as well as creating agents to perform certain routine tasks.

“The idea is that we want people to have this unified interface from which you can answer questions, do research, take actions, all of that,” Jose Kunnackal John, director of Amazon Quick Suite, said.

The move positions AWS to compete with the likes of 

, which have their own agentic tools for enterprise. These companies and others have 

 as the next chapter of generative AI adoption in the workplace.

What sets Amazon apart, Kunnackal John said, is that AWS customers wanted an AI system that can operate across the many apps and platforms that workers use in a typical day.

“There is this plethora of tools in the enterprise, and every tool has AI,” he said. “Google’s got its own suite. Microsoft’s got its own suite. You can pick every provider; they have their own suite of tools. The idea is, can you bring all of the context that you have, all of the data that you have, into one place? And often that’s not the case, and with Quick Suite, we are trying to make that the case.”

 At Amazon’s offices in Manhattan, the company offered a demo of Quick Suite focused on different jobs—IT, marketing, sales, journalism (the last presumably more of a play to the audience in the room than a big potential business line).

You can ask questions or generate longer research reports drawn from a variety of different information sources—internal reference docs, data warehouses, and the open web. You can create actions—or Quick Flows—that will automatically do things like update certain tickets or help onboard new workers into systems. It can send messages or recap emails.

For the media pros in the room, Kunnackal John suggested providing the system with past comments from an editor to train it to anticipate future critiques on story drafts. Or maybe to generate a research overview on existing news stories about a given topic.

 for building AI agents this summer. But as a workplace software-as-a-service (SaaS) platform, Quick Suite builds more on Amazon Q, Amazon’s enterprise AI assistant, which has faced accuracy issues in the past, according to a 

 last week that cited internal review documents.

Quick Suite has been beta tested with hundreds of customers across healthcare, manufacturing, and travel, including 3M, Vertiv, and Jabil, as well as tens of thousands of Amazon employees, according to the announcement.

“People are creating chat agents and flows that essentially take away that heavy lifting that they’ve had to do, and that eases their work,” Kunnackal John said. “It’s not that I’m massively rolling out an AI project. AI is not something scary. This makes it easy for them to use it and to adopt it in their own workflows, whatever that is.”

AI