If it feels these days as if everything in technology is about AI, that’s because it is. And nowhere is that more true than in the market for computer memory. Demand, and profitability, for the type of DRAM used to feed GPUs and other accelerators in AI data centers is so huge that it’s diverting away supply of memory for other uses and causing prices to skyrocket. According to Counterpoint Research, DRAM prices have risen 80-90 precent so far this quarter.

The largest AI hardware companies say they have secured their chips out as far as 2028, but that leaves everybody else—makers of PCs, consumer gizmos, and everything else that needs to temporarily store a billion bits—scrambling to deal with scarce supply and inflated prices.

How did the electronics industry get into this mess, and more importantly, how will it get out? IEEE Spectrum asked economists and memory experts to explain. They say today’s situation is the result of a collision between the DRAM industry’s historic boom and bust cycle and an AI hardware infrastructure build-out that’s without precedent in its scale. And, barring some major collapse in the AI sector, it will take years for new capacity and new technology to bring supply in line with demand. Prices might stay high even then.

To understand both ends of the tale, you need to know the main culprit in the supply and demand swing, high-bandwidth memory, or HBM.

What is HBM?

HBM is the DRAM industry’s attempt to short-circuit the slowing pace of Moore’s Law by using 3D chip packaging technology. Each HBM chip is made up of as many as 12 thinned-down DRAM chips called dies. Each die contains a number of vertical connections called through silicon vias (TSVs). The dies are piled atop each other and connected by arrays of microscopic solder balls aligned to the TSVs. This DRAM tower—well, at about 750 micrometers thick, it’s more of a brutalist office-block than a tower—is then stacked atop what’s called the base die, which shuttles bits between the memory dies and the processor.

This complex piece of technology is then set within a millimeter of a GPU or other AI accelerator, to which it is linked by as many as 2,048 micrometer-scale connections. HBMs are attached on two sides of the processor, and the GPU and memory are packaged together as a single unit.

The idea behind such a tight, highly-connected squeeze with the GPU is to knock down what’s called the memory wall. That’s the barrier in energy and time of bringing the terabytes per second of data needed to run large language models into the GPU. Memory bandwidth is a key limiter to how fast LLMs can run.

As a technology, HBM has been around for more than 10 years, and DRAM makers have been busy boosting its capability.

As the size of AI models has grown, so has HBM’s importance to the GPU. But that’s come at a cost. SemiAnalysis estimates that HBM generally costs three times as much as other types of memory and constitutes 50 percent or more of the cost of the packaged GPU.

Origins of the memory chip shortage

Memory and storage industry watchers agree that DRAM is a highly cyclical industry with huge booms and devastating busts. With new fabs costing US $15 billion or more, firms are extremely reluctant to expand and may only have the cash to do so during boom times, explains Thomas Coughlin, a storage and memory expert and president of Coughlin Associates. But building such a fab and getting it up and running can take 18 months or more, practically ensuring that new capacity arrives well past the initial surge in demand, flooding the market and depressing prices.

The origins of today’s cycle, says Coughlin, go all the way back to the chip supply panic surrounding the COVID-19 pandemic . To avoid supply-chain stumbles and support the rapid shift to remote work, hyperscalers—data center giants like Amazon, Google, and Microsoft—bought up huge inventories of memory and storage, boosting prices, he notes.

But then supply became more regular and data center expansion fell off in 2022, causing memory and storage prices to plummet. This recession continued into 2023, and even resulted in big memory and storage companies such as Samsung cutting production by 50 percent to try and keep prices from going below the costs of manufacturing, says Coughlin. It was a rare and fairly desperate move, because companies typically have to run plants at full capacity just to earn back their value.

After a recovery began in late 2023, “all the memory and storage companies were very wary of increasing their production capacity again,” says Coughlin. “Thus there was little or no investment in new production capacity in 2024 and through most of 2025.”

The AI data center boom

That lack of new investment is colliding headlong with a huge boost in demand from new data centers. Globally, there are nearly 2,000 new data centers either planned or under construction right now, according to Data Center Map. If they’re all built, it would represent a 20 percent jump in the global supply, which stands at around 9,000 facilities now.

If the current build-out continues at pace, McKinsey predicts companies will spend $7 trillion by 2030, with the bulk of that—$5.2 trillion—going to AI-focused data centers. Of that chunk, $3.3 billion will go toward servers, data storage, and network equipment, the firm predicts.

The biggest beneficiary so far of the AI data center boom is unquestionably GPU-maker Nvidia. Revenue for its data center business went from barely a billion in the final quarter of 2019 to $51 billion in the quarter that ended in October 2025. Over this period, its server GPUs have demanded not just more and more gigabytes of DRAM but an increasing number of DRAM chips. The recently released B300 uses eight HBM chips, each of which is a stack of 12 DRAM dies. Competitors’ use of HBM has largely mirrored Nvidia’s. AMD’s MI350 GPU, for example, also uses eight, 12-die chips.

With so much demand, an increasing fraction of the revenue for DRAM makers comes from HBM. Micron—the number three producer behind SK Hynix and Samsung—reported that HBM and other cloud-related memory went from being 17 percent of its DRAM revenue in 2023 to nearly 50 percent in 2025.

Micron predicts the total market for HBM will grow from $35 billion in 2025 to $100 billion by 2028—a figure larger than the entire DRAM market in 2024, CEO Sanjay Mehrotra told analysts in December. It’s reaching that figure two years earlier than Micron had previously expected. Across the industry, demand will outstrip supply “substantially… for the foreseeable future,” he said.

Future DRAM supply and technology

“There are two ways to address supply issues with DRAM: with innovation or with building more fabs,” explains Mina Kim, an economist with the Mkecon Insights. “As DRAM scaling has become more difficult, the industry has turned to advanced packaging… which is just using more DRAM.”

Micron, Samsung, and SK Hynix combined make up the vast majority of the memory and storage markets, and all three have new fabs and facilities in the works. However, these are unlikely to contribute meaningfully to bringing down prices.

Micron is in the process of building an HBM fab in Singapore that should be in production in 2027. And it is retooling a fab it purchased from PSMC in Taiwan that will begin production in the second half of 2027. Last month, Micron broke ground on what will be a DRAM fab complex in Onondaga County, N.Y. It will not be in full production until 2030.

Samsung plans to start producing at a new plant in Pyeongtaek, South Korea in 2028.

SK Hynix is building HBM and packaging facilities in West Lafayette, Indiana set to begin production by the end of 2028, and an HBM fab it’s building in Cheongju should be complete in 2027.

Speaking of his sense of the DRAM market, Intel CEO Lip-Bu Tan told attendees at the Cisco AI Summit last week: “There’s no relief until 2028.”

With these expansions unable to contribute for several years, other factors will be needed to increase supply. “Relief will come from a combination of incremental capacity expansions by existing DRAM leaders, yield improvements in advanced packaging, and a broader diversification of supply chains,” says Shawn DuBravac , chief economist for the Global Electronics Association (formerly the IPC). “New fabs will help at the margin, but the faster gains will come from process learning, better [DRAM] stacking efficiency, and tighter coordination between memory suppliers and AI chip designers.”

So, will prices come down once some of these new plants come on line? Don’t bet on it. “In general, economists find that prices come down much more slowly and reluctantly than they go up. DRAM today is unlikely to be an exception to this general observation, especially given the insatiable demand for compute,” says Kim.

In the meantime, technologies are in the works that could make HBM an even bigger consumer of silicon. The standard for HBM4 can accommodate 16 stacked DRAM dies, even though today’s chips only use 12 dies. Getting to 16 has a lot to do with the chip stacking technology. Conducting heat through the HBM “layer cake” of silicon, solder, and support material is a key limiter to going higher and in repositioning HBM inside the package to get even more bandwidth.

SK Hynix claims a heat conduction advantage through a manufacturing process called advanced MR-MUF (mass reflow molded underfill). Further out, an alternative chip stacking technology called hybrid bonding could help heat conduction by reducing the die-to-die vertical distance essentially to zero. In 2024, researchers at Samsung proved they could produce a 16-high stack with hybrid bonding, and they suggested that 20 dies was not out of reach.

Presented by EdgeVerve

Before addressing Global Business Services (GBS), let’s take a step back. Can agentic AI, the type of AI able to take goal-driven action, transform not just GBS but any kind of enterprise? And has it done so yet?

As with many new technologies, rhetoric has outpaced deployment in this case. While 2025 was “supposed to be the year of agentic AI,” it didn’t turn out that way, according to VentureBeat Contributing Editor Taryn Plumb. Leaning on input from Google Cloud and integrated development environment (IDE) company Replit, Plumb reported in a December 2025 VentureBeat post that what has been missing are the fundamentals required to scale.

Given the experience of Large Language Model (LLM)-based generative (gen)AI, this outcome is not surprising. In a survey conducted at the February 2025 Shared Services & Outsourcing Network (SSON) summit, 65% of GBS organizations responded that they had yet to complete a GenAI project. One can safely say that the adoption of the more recently arrived agentic AI is still in its very nascent stages for enterprises, including GBS.

The role of agentic AI in Global Business Services

There are good reasons, nonetheless, to focus on the tremendous potential of agentic AI and its application to the GBS sector.

Stripped of hype, Agentic AI unlocks capabilities in the orchestration layer of software workflows that weren’t practical before. It does so through a range of techniques, including (but not requiring) LLMs. While enterprises may indeed be missing certain fundamentals needed to deploy agentic AI at scale, those prerequisites are not out of reach.

As for GBS and Global Capability Centers (GCCs), they have already been undergoing a makeover, from back-office extensions into increasingly strategic enterprise partners. Agentic AI is a natural fit because one of its standard use cases involves IT operations or customer-service agents, functionality already within the existing GBS and GCC wheelhouse.

So yes, agentic AI could potentially transform the GBS sector. Industry leaders can best move toward scaled deployment by taking a methodical approach.

Five steps for deploying agentic AI in GBS

Agentic AI is not the only game in town. As noted, there’s GenAI, used primarily for content creation. But broadening the scope, we can also point to predictive AI and document AI, used respectively for forecasting and data extraction. (Neither requires LLMs.) Exposure to preexisting AI bodes well for the future of agentic AI.

First, these flavors of AI are mutually supportive, stacked (rather than siloed) in modern systems. Agentic AI, in particular, is positioned to draw upon the others. Second, having lived through the hype cycle of GenAI, industry leaders may be inclined to take a more measured – and productive – approach to agentic AI.

Rather than rushing into a pilot, the industry would do well to prep carefully (steps 1-3). When combined with the right test project (step 4), these actions can pave the way for a scaled-up deployment of agentic AI (step 5):

Know thy processes. Business operations can be complicated. Consider a top global shipping and logistics firm, whose thousands of full-time employees at its seven GBS centers supported more than 80 processes involving highly complex, manually intensive workflows with wide regional variations. Only by first understanding existing processes and workflows does an organization like this stand a chance of being able to rethink or rework them.

Know thy data. Closely related are the data that workflows depend upon. How do these data flow from end to end? What do the pipelines look like? Where are the key APIs? Are the data structured or unstructured? Do the resources include data platforms (systems of record) and vector databases (context engines), both of which AI agents need to make good decisions? What kind of data governance and security prevail? How might those change in an agentic AI scenario?

Identify the problem. In the case of the shipping firm mentioned above, the complexity and variation of the workflows, as well as their manual intensity, exposed it to significant costs, lapses in service level agreements (SLAs), poor customer experience and heightened compliance and legal risks. Once named, a problem logically becomes a potential use case with discrete objectives.

Pilot an operating model. Options include consolidating efforts in a Center of Excellence (COE), democratizing development through citizen-led approaches, and partnering through Build-Operate-Transform-Transform-Transfer (BOTT) models, among others. Without structural clarity, even promising AI pilots are difficult to extend beyond their initial domain. The model should also reflect reality. Likely involving multiple, parallel agents in pursuit of coordinated goals, Agentic AI is still constrained by environment, complexity, risks and governance.

Scale up. Successful pilots lead to their own next steps. Take the fragmented experience of a large multinational bank in Australia. After automating several non-core processes through Automation COE, the bank realized it needed to analyze and improve its most complex workflows. It selected an over-the-top software platform that enabled it to complete more than 100 discovery projects in under 14 months. Pilots thus may grow, becoming enterprise-wide initiatives.

What agentic AI looks like at enterprise scale

Only scale can yield real impact. The shipping provider, with its seven GBS centers, ended up with technology capable of building data pipelines, digitizing complex documents, applying rule-based reasoning across country-specific exceptions and orchestrating work across teams. That foundation led to an AI-first transformation of about 16 initiatives, exponential growth in automation and significant efficiency gains.

By unleashing capabilities at the orchestration layer – enabling contextual perception, cross-domain collaboration, and autonomous action aligned with governance – agentic AI can turbo-charge operations, both AI and human.

Consider a procurement process. While document AI can extract data from purchase orders, obviating certain manual checks, an AI agent could also evaluate vendor risk, cross-reference compliance standards, verify budget availability and even initiate negotiation while keeping audit logs for regulatory reporting. In a financial advisory scenario, while predictive AI can analyze trends, an AI agent could take further action, assisting professionals in particular business units on targeted strategic investments.

Note that the agent isn’t replacing human judgment, but extending it, ensuring decisions are made faster, more consistently and on a scale.

From standalone automation to agentic ecosystems in GBS

GBS is uniquely positioned to lead the enterprise into the agentic AI era. By design, GBS sits at the intersection of processes and data across multiple business units. Finance, HR, supply chain and IT all flow through the shared services model. This central vantage point makes GBS an ideal launchpad for creating agentic AI ecosystems.

An ecosystem differs from standalone automation. Agents don’t perform tasks in isolation. Rather, they work as part of an interconnected system. They share insights, learn from one another and coordinate to optimize outcomes at the enterprise level. Deployed within a GBS or GCC, Agentic AI can accelerate their ongoing transformation, enabling them to leapfrog incremental automation and operate at the level of end-to-end process orchestration.

N. Shashidar is SVP & Global Head, Product Management at EdgeVerve.

Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact sales@venturebeat.com.

Instead of reinventing the wheel, Hong Kong dim sum chain Tim Ho Wan is doubling down on the basics

Eateries have had a tough couple of years. In 2024, Singapore saw its highest number of F&B business closures in 20 years, with over 3,000 establishments shutting their doors. 2025 witnessed a similar trend, with the Ministry of Trade and Industry reporting that 2,431 businesses closed in the first 10 months of the year. 

Apart from local operators, international chains have also shrunk and even succumbed to this rapid pace of closures, with the likes of US’ Eggslut, England’s PizzaExpress and Taiwan’s Hollin proving that scale and recognition are no longer a safety net in Singapore’s F&B scene.

So when Vulcan Post received an exclusive invitation to a media tour and witness the opening of Tim Ho Wan’s tenth store in Hong Kong last week, it was admittedly easy to dismiss the offer as “another store opening.” However, through the tour, we found out how their strategy might be the right counter to Singapore’s tough F&B scene: starting with introducing new products that pay homage to their classics.

Vulcan Post caught up with Yeong Sheng Lee, CEO of Tim Ho Wan, in an exclusive interview to understand how the business plans to continue thriving in Singapore’s cutthroat F&B scene, starting with their products. 

Why Singapore is a tough nut to crack for international brands

Having managed multiple F&B brands in Southeast Asia for almost a decade, Yeong Sheng Lee knows just how challenging the food scene really is, especially in Singapore. 

While Singapore presents itself as an open market for foreign brands, multiple factors can result in costly miscalculations that make even the biggest of brands struggle. Lee acknowledged that the market has seen an increase in challenges post COVID-19, including rising operational costs, a shortage of labour, and meeting customers who have high standards for both food quality and value. 

“What I actually mean by value is that not everyone is looking for things that are cheap, and there’s a clear difference. Cheap is when I always want to go for the lowest pricing, but value is where for the price that I’m paying, am I getting more than what I should actually be expecting?”  

When asked if the pessimism shared by Singaporean netizens and entrepreneurs is impeding the local market, Lee remained optimistic that the brand would continue to grow its presence in the country for the long term rather than pulling back like many other chains. 

“I wouldn’t actually say it’s pessimism. It’s just a phase that we actually go through, like every other market would actually have its highs and lows. No doubt, while it is actually challenging in Singapore for our team, we remain optimistic and confident about the Singapore market.” 

But what is the secret to his confidence? It turns out to be simpler than one might expect. 

Redefining Value for the Singaporean Market 

As the food industry is an extremely trend-sensitive sector, Lee emphasised the importance of constant innovation to keep up with short attention spans, and this comes from introducing seasonal items, from limited-time-only menus to elevated versions of bestsellers. 

For example, the success of Tim Ho Wan’s signature pork bun gave inspiration to a new item. Lee shared that the chefs started working to create a cousin to their signature baked pork bun, and were exploring ingredients that could create a black-coloured crust. They decided on truffle, following the success of truffle food offerings worldwide.

“For the price that I’m paying, I’m getting a black truffle chicken bun, something which would be a strong indication of value to a lot of consumers. So value in this case is what are we actually putting into our products? What are the ingredients we’re using?” 

This ethos also translates into their festive menus. For this year’s Chinese New Year menu, Tim Ho Wan introduced limited edition dishes, such as steamed scallops—an ingredient not often found in fast casual chains—deep-fried nian gao (rice cake) and steamed chicken in wine broth. Localised offerings, including Bak Kut Teh dumplings and Musang Wang durian sesame balls, also help to maintain relevance in the local market.

Lee also added that chefs play a major role in menu innovations, and they are constantly experimenting with new seasonal products based on on-hand customer research. This allows them to avoid introducing products that do not resonate with the local crowd, reducing risk in a market where failed launches are costly.  

Beyond menu innovations, operations also matter. Lee pointed out that food deliveries have been growing in Singapore, and the business has been working on curating its menu with products—including bento sets—that are more durable for deliveries. Tim Ho Wan has been introducing refreshed layouts of their stores, such as its Westgate outlet, with more open layouts and brighter lighting.

While these initiatives seem mundane at first, they are also crucial for the brand’s survival in a time when many restaurants have been experiencing declining traffic. 

Ensuring consistency even as the business scales up

As the brand continued to elevate its offerings, consistency was another factor that Tim Ho Wan had to get right, especially after the brand was acquired by Jollibee Group in Jan 2025.

Instead of giving Tim Ho Wan a total overhaul, the first thing Lee focused on was building upon the brand’s strengths, working with the chefs and backend operation teams to understand how the brand grew and thrived over a decade: from getting their first Michelin Star a year after opening, to expanding to 11 markets in Asia, Australia and the US.

Leveraging on Jollibee’s expertise, Lee helped the company streamline their operations across company-owned establishments, joint ventures and franchises. He also introduced the Experts Chef Program: a training programme where chefs across all markets will come to Hong Kong to learn Tim Ho Wan’s recipes and replicate them to the same standard. 

To verify this on the ground, Vulcan Post also spoke to Chef Cheung Yit Sing, who is also Global Product Innovation Director of Tim Ho Wan, to learn more about the training process and how the business foundation is kept.  

From working at pushcart restaurants to cooking at hotels and conventions, Chef Cheung has been making dim sum since 1982 and is now in charge of teaching Tim Ho Wan’s recipes to other chefs. He shared that as the creator of a specific dish, he starts every training with a lecture, followed by a demonstration of how it’s made. 

The chefs will then get to taste the dish to understand the standard that they need to hit before making it. After learning the recipes, the chefs will then undergo an assessment to be scored for them to maintain their operational benchmarks. 

“Currently, our passing mark is at 90 out of 100, because the standard is really high to ensure that when they are able to match the quality when they leave,” Chef Cheung shared in Mandarin, adding that it will take at least a day for the chefs to learn two recipes. 

Don’t fix what doesn’t need to be fixed 

Globally, Tim Ho Wan has experienced a 5.2% increase in revenue in Q3 2025 compared to the first half of the year, with encouraging results in newer market entries in the United States. Following the opening of their first company-owned outlet in Irvine, California, Lee shared that the country will be a key market for the brand and expressed his ambition to open 20 additional stores by 2028.

In Singapore, Tim Ho Wan saw a 7% growth in the same period, and the brand has nine outlets in the country since its entry 13 years ago. While Lee revealed that the company is constantly looking for new opportunities to introduce more locations, he also emphasised that the team is not rushing to open new locations to meet timelines—a move that differs from other big chains that overstretched after aggressively expanding.  

“We will continue looking at opportunities, but at the same time, we are actually investing in building our organisation’s capability and strengthening our systems and processes,” he explained. “It’s a combination of these factors that we continue to look at how we accelerate our growth in Singapore sustainably, because we are not in this for the short term.”  

In a market where even well-known international names are struggling to stay afloat, Tim Ho Wan’s strategy stands in contrast to the churn. Rather than expanding aggressively or reinventing its identity, the brand focused on offering classics and modern versions of dim sum, while also staying intentional with its expansions.

• Learn more about Tim Ho Wan here.
• Read more stories we’ve written on Singaporean businesses here.

Featured Image Credit: Tim Ho Wan

In August 2019, I walked into an early learning center in Philadelphia with a blank reporter’s notebook, a camera and a whole lot to learn.

Prior to that, I’d covered K-12 education and a bit of higher ed. The worlds of child care and early childhood education were foreign to me. I didn’t know the lingo or the layout. And, as I’d learn moments later, I didn’t have the slightest idea what it entailed to care for and engage babies and toddlers.

Fast forward six years, and now my favorite part of my work as an education journalist is meeting and writing about young children and their caregivers.

It wasn’t long after that program visit in Philly that I began to feel this way. Back then, few news outlets consistently published stories about the early years. EdSurge won grant funding to cover early childhood, and because no one on staff had covered the field before, we were given a sizable budget for travel. The idea was that I’d learn the beat in context. I went to early care and education programs all over the country — in homes, centers, schools and churches. I saw what an early learning environment looked like. I heard the sounds — oh, the glorious sounds of laughter and squeals of delight, the tears of a toy snagged unjustly or an unwanted nap. I smelled the smells. I noted the physicality of the job. I watched closely enough to realize that, as the children played — whether inside or outside, independently or in groups, structured or unstructured, real or imaginative — they were developing critical life skills.

From Utah to Ohio to New Jersey, I was filled with wonder during those initial months on the early childhood beat. I loved watching the way young children think and move about the world. I could not believe the depths of patience their teachers had. I puzzled over how, despite everything I learned about the significance of the early years for brain development and long-term success, child care was sorely underfunded, leaving families, educators and kids to figure it out for themselves.

It is one thing to write about babies’ brain development and skill acquisition, to cover the backwardness of the U.S. early care and education system, to report on the impossible choices parents are asked to make. It’s another to live it.

When I became pregnant with my first child in 2024, I told my husband that, as soon as we heard the baby’s heartbeat, I wanted to begin our child care search.

We hadn’t even told most friends and family when we began touring early learning centers in Denver. I expected to join long waitlists. I expected it would become our biggest or second-biggest expense, after housing. I’d been writing about these realities for years, after all.

But even I was shocked to be told, by more than one program director, that they likely wouldn’t have a slot for our son — who was due in spring 2025 — until 2027 or 2028.

And when we eventually decided to pursue a nanny share — in which our child and another child receive care from the same nanny in one of the family’s homes — I was prepared for a high-stakes hiring process. But I didn’t realize, until I got into it, just how difficult it would be to find someone with whom I felt I could entrust the single most precious thing in my life. Or how conflicted I would feel to be at my desk, writing about other child care arrangements for other people’s kids, when I could hear my own baby laughing and crying and babbling right upstairs.

Then there is the baby himself.

I think back to what I didn’t know and what I assumed back in 2019, and I shake my head. Little kids don’t just come online one day, around age 4 or 5, even though that’s how the education system in America treats them.

Some weeks into his life, I watched my son discover his hands. And then I watched him use those hands to reach for a bell that hung over his playmat. After he figured out how to touch it, he learned to grasp it, and after he learned to grasp it, he mastered ringing it. Now, at 7 months old, he uses those hands to pick up board books and hold camping mugs and shake rattles and grab my face. He picks up foods like crusty bread and roasted carrots and strips of scrambled egg and brings them to his mouth to eat. I marvel.

I’ve heard experts explain for years that close caregiver relationships are what a child needs most in the first year of life. But in recent months, I’ve come to see firsthand how much comfort and encouragement and joy mine and my husband’s presence provide our son. I see him look to us for reactions. Now that he’s crawling, he follows us from room to room. Now that he’s reaching, I know when he wants to be held. Now that he’s been in a nanny share for some time, I know that he’s built a relationship with the nanny because he lights up when she arrives for the day.

I can’t say for certain that early childhood reporting has made me a better mom. Perhaps, in subtle ways, some kernels of knowledge have carried over. But I feel quite sure — at the very least hopeful — that motherhood will make me a more perceptive reporter, keenly aware of the stakes in early childhood and more empathetic to those the field touches.

On that subject, this is my last piece as a senior reporter for EdSurge. It has been a great run, with nearly 300 published stories over more than seven years. I’ve covered K-12 and early childhood education here with enthusiasm and commitment, even amid company mergers, a global pandemic, layoffs, and many seasons of change in my personal life.

The early childhood beat has grown up a bit in that time too, with many major newsrooms now devoting full-time positions to the field.

EdSurge will continue to cover early care and education after my time here is up. And in my next chapter as a journalist, so will I. I expect our paths will cross again and again.

• Control Resonant does not use generative AI content “at all”
• Remedy interim CEO Markus Mäki confirmed that there is “varied interest” in the tech at the company
• He adds that the company is following AI progression to see if anything can be used “ethically” that would “add player value”

Remedy interim CEO Markus Mäki has confirmed that the studio’s next major title, Control Resonant, does not utilize generative AI. However, he admits that there is interest in the tech at the company.

Speaking during Remedy’s most recent earnings Q&A (via Game Developer), Mäki discussed the divisive technology that has been seeping into every facet of the games industry over the past few years.

The fake goods crisis cuts two ways. Luxury brands lose more than $30 billion a year to counterfeits, while buyers in the booming $210 billion second-hand market have no reliable way to verify that what they’re purchasing is genuine. Veritas wants to solve both problems with a solution that combines custom hardware and software.

The startup claims that it has developed a “hack-proof” chip that can’t be bypassed by devices like Flipper Zero, a widely available hacking tool that can be used to tamper with wireless systems. These chips are linked with digital certificates to verify the authenticity of the products.

Vertitas founder Luci Holland has experienced life as both a technologist and an artist. She has worked in different artistic mediums, including mixed media painting and metal sculpture. She has also worked at Tesla as a technical product manager and has held several business development, community growth, and product management roles at tech companies and venture funds.

Holland noted that traditionally, luxury goods makers use various symbols or physical marks to authenticate their products. However, with the growing demand for these goods, counterfeiters have learned to create convincing copies of these marks along with high-quality fake certificates. These goods are often called “superfakes.”

Holland mentioned that she spoke with maisons — established luxury fashion houses — that said that some of their locations had to stop authenticating goods because fakes were becoming too convincing to reliably detect. She said that drawing on her experience in both the tech and art worlds, she wanted to solve the problem.

“For me, as someone who has a background in being a designer and then also has experience in tech, I saw this problem and thought about the different ways we could solve it. I think what’s truly innovative is we’ve used and combined elements from both hardware and software to create this solution that helps protect brands in this way to convey the information,” she said.

“When I think of counterfeiting and I think of the most iconic and legacy brands,” she added, “a lot of these brands have been around for over 100, 150 years. These brands deserve the most advanced protection to protect these designs.”

Veritas worked with different designers to create a chip that is minimally disruptive to the product creation process. The chip is the size of a small gem and can easily be inserted even after a product is made without compromising its integrity. The chip incorporates NFC, or Near Field Communication — the same short-range wireless technology used in contactless payments. This means you can tap your smartphone on the item to verify its authenticity.

Holland said that for security purposes, the startup developed a custom coil and a bridge structure. If someone attempts to tamper with the product, the chip goes dormant and hides the codes related to the product. On the software side, the product information is linked to the Veritas back end, which monitors scanning behavior to prevent fraud. The company also creates a blockchain-based digital clone of the product for possible digital art gallery shows or metaverse activities.

The company didn’t reveal who it is working with, but said that brands can use its software suite to get information about all the chipped products, add team members to manage items and add product information along with the product story — details that can also be used to connect with their community. The startup said that some partners use this to engage customers through exclusive invitations or early access to new products.

While the counterfeiting market is big, Holland thinks the market still needs education around why it needs robust tech solutions.

“It is shocking to see that some of the shelf solutions, like NFC chips that brands are using, are actually so vulnerable and could easily be bypassed. This is the one thing most people don’t know, and we want to educate the ecosystem to adopt safer solutions,” Holland said.

Veritas said that it raised $1.75 million in pre-seed funding led by Seven Seven Six, along with Doordash co-founder Stanley Tang, skincare brand Reys’ co-founder Gloria Zhu, and former TechCrunch editor Josh Constine. The company plans to use the funding to expand its two-person team.

Seven Seven Six’s Alexis Ohanian said that he was impressed by the combination of design taste and technology expertise of Holland. He thinks that brands know that fake goods are a problem and are constantly looking for robust solutions.

“It’s absolutely an arms race [against fake goods makers], but we’re used to fighting those and consistently winning in tech — and luxury brands need all the help they can get,” Ohanian said.