Techin' Geek

TikTok, which is managed by the Chinese firm ByteDance, has found itself embroiled in controversy in the U.S. for the past four years owing to fears regarding user data possibly being accessed by the Chinese authorities.

Consequently, users in the U.S. have frequently felt trapped within this conflict. Earlier this year, the app faced a brief outage in the U.S. that left millions of users on edge before it was swiftly restored. In February, TikTok made its return to both the App Store and Google Play Store. 

Several investors vied to acquire the app, and with Trump having extended the TikTok ban deadline for the fourth time, the competition has finally reached its conclusion. As of last week, TikTok has officially sealed a deal to divest a segment of its U.S. operations to a coalition of American investors.

This agreement comes almost three months after President Donald Trump enacted an executive order approving the sale of TikTok’s U.S. operations to a group of American investors.

Just a week earlier, President Trump revealed that President Xi Jinping of China had granted his endorsement of a TikTok deal, enabling a consortium of U.S. investors to manage the platform. ByteDance publicly assured it would maintain the platform’s accessibility for American users.

Who possesses TikTok in the U.S.?

As per a memo obtained by TechCrunch, the investor group includes Oracle, the private equity firm Silver Lake, and the investment firm MGX. Together, they will possess 45% of the U.S. operation, while ByteDance will retain approximately a 20% ownership. Axios was the first to report the information, citing sources estimating TikTok U.S. to be valued at around $14 billion — a valuation also noted by Vice President JD Vance.

In September, reports indicated that a “framework” deal had been established between the U.S. and China, involving a consortium of investors — including Oracle, Silver Lake, and Andreessen Horowitz — overseeing TikTok’s U.S. operations. These investors were anticipated to hold an 80% stake, while the remaining shares would be owned by Chinese stakeholders.

The newly established “TikTok USDS Joint Venture LLC” will manage the app’s operations, which include data security, algorithm integrity, content oversight, and software reliability.

According to the memo, Oracle will act as the designated security partner, charged with auditing and ensuring adherence to National Security Terms. The company already supplies cloud solutions for TikTok and handles user data within the U.S. Notably, Oracle had previously attempted to acquire TikTok in 2020.

A White House official had earlier stated that Oracle would duplicate and secure a new U.S. version of the algorithm, allowing U.S.-based TikTok proprietors to lease the algorithm from ByteDance, which Oracle will then retrain. 

ByteDance will not have access to data regarding TikTok’s U.S. users or any control over the U.S. algorithm.

The transaction is set to be finalized by January 22, 2026.

What U.S. users ought to know

Reports from Bloomberg suggest that once the agreement is finalized, the TikTok app will be phased out in the U.S., and users will need to migrate to a different platform. Nonetheless, the details of this platform are predominantly uncertain, encompassing its features and how it will diverge from the original app. 

How did we arrive at this point?

To grasp the full scope of this high-stakes saga, we must first revisit the timeline of TikTok’s contentious relationship with the U.S. government, which led to numerous legal battles and negotiations. 

The saga initially unfolded in August 2020, when Trump issued an executive order to restrict transactions with the parent company ByteDance. 

A month later, Trump’s administration endeavored to compel a sale of TikTok’s U.S. operations to an American company. The primary contenders were Microsoft, Oracle, and Walmart. However, a U.S. judge temporarily halted Trump’s executive order, enabling TikTok to keep operating as the legal proceedings continued. 

Progress resumed last year with the transition to the Biden administration. Following the Senate’s passage of a bill against TikTok, President Joe Biden signed it.

In retaliation, TikTok filed a lawsuit against the U.S. government, contesting the legality of the ban and claiming that the app and its American users were having their First Amendment rights breached. The company has consistently refuted the notion that it poses a security risk, maintaining that its data stored in the U.S. adheres to all local regulations.

Fast-forward to now: Trump has shifted his stance since his initial term and is pursuing a 50-50 ownership split between ByteDance and a U.S. company. 

Various contenders have emerged, including The People’s Bid for TikTok, a consortium led by Project Liberty founder Frank McCourt. This group is backed by investment firm Guggenheim Securities and the law firm Kirkland & Ellis. Supporters encompass Reddit co-founder Alexis Ohanian, TV figure and investor Kevin O’Leary, World Wide Web inventor Tim Berners-Lee, and senior research scientist David Clark.

Another competing group, the American Investor Consortium, is spearheaded by Employer.com founder Jesse Tinsley and includes Roblox co-founder David Baszucki, Anchorage Digital co-founder Nathan McCauley, and renowned YouTuber MrBeast.

Additional contenders have included Amazon, AppLovin, Microsoft, Perplexity AI, Rumble, Walmart, Zoop, former Activision CEO Bobby Kotick, and former U.S. Treasury Secretary Steven Mnuchin.

The narrative has been updated following publication.

Mark Zuckerberg has made another move.

Meta Platforms is set to acquire Manus, an AI startup based in Singapore that has generated significant buzz in Silicon Valley since its launch last spring with a demo showcasing an AI agent performing tasks like screening job applicants, organizing vacations, and evaluating stock portfolios. At the time, Manus claimed it surpassed the capabilities of OpenAI’s Deep Research.

In April, shortly after its debut, venture capital firm Benchmark spearheaded a $75 million funding round that valued Manus at $500 million post-money, with Benchmark general partner Chetan Puttagunta joining Manus’ board. Reports from Chinese news sources indicated that other prominent investors had already backed Manus, including Tencent, ZhenFund, and HSG (formerly Sequoia China) with a $10 million investment.

The company revealed in mid-December that it has acquired millions of users and is producing annual recurring revenue exceeding $100 million from its subscription service.

Around this time, Meta began discussions with Manus, according to the WSJ, which reports that the tech giant is paying $2 billion — the valuation Manus was reportedly pursuing for its upcoming funding round.

For Zuckerberg, who has aligned Meta’s vision with AI, Manus signifies a new opportunity: an AI product that is already profitable. This is particularly relevant as investors have become increasingly anxious about Meta’s $60 billion infrastructure investment and the tech sector’s debt-financed outlays for data center development.

Meta has stated that it will allow Manus to operate independently while integrating the startup’s AI agents into Facebook, Instagram, and WhatsApp, where Meta’s own chatbot, Meta AI, is already accessible to users.

There is, however, an issue: Manus’ founders, originating from China, established its parent company, Butterfly Effect, in Beijing in 2022 before relocating to Singapore in the middle of this year. Whether this will raise concerns in Washington is yet to be determined, but Senator John Cornyn has already criticized Benchmark for its investment in the firm, echoing worries in May about American funds being funneled to a Chinese entity.

Cornyn, a Republican from Texas and a senior figure on the Senate Intelligence Committee, has consistently been one of Congress’ most outspoken advocates regarding China and tech competition, though he is not alone. Adopting a tough stance against China has emerged as one of the few truly bipartisan topics in Congress.

It comes as no surprise that Meta has informed Nikkei Asia that post-acquisition, Manus will sever its connections with Chinese financiers and will cease operations in China. “There will be no ongoing Chinese ownership interests in Manus AI following the transaction, and Manus AI will halt its services and operations in China,” a Meta spokesperson conveyed to the outlet.

At TechCrunch Disrupt, three investors stepped onto the stage to analyze the elements that can either enhance or undermine a pitch deck. Jyoti Bansal, a founder-turned-investor; Medha Agarwal from Defy; and Jennifer Neundorfer of January Ventures conveyed their honest opinions to the audience regarding what is effective in a pitch deck and what isn’t.

Their main frustration? Overuse of buzzwords.

According to Agarwal, the more frequently a founder mentions AI in their pitch, the less likely it is that the company effectively utilizes AI. “Those who are genuinely innovating will reference it, and it’s integrated into their work, but it’s not the focal point of their pitch,” she explained to the attendees.

Bansal, who has created and sold several businesses before transitioning to investment, summarized what investors expect into three essential questions. First, he inquires whether there’s a sufficiently large market to address. Does the founder’s concept have the capability to evolve into a significant enterprise? And is the issue they are addressing genuinely worth addressing?

Secondly, investors seek to understand why this particular founder is the right person to lead the company. “There must be something distinctive about you,” Bansal noted, mentioning that this could involve possessing exceptional team members or unique skills. “What makes you the victor? If the problem is compelling, there will likely be numerous other companies attempting to resolve it, so what makes you the one to succeed and what’s your edge?”

The third aspect investors look for, Bansal mentioned, is some form of validation. “Customer traction,” he stated. “Validation could stem from initial customer input, revenue, or another form, but some sort of validation is crucial.”

These three inquiries, Bansal emphasized, ultimately lead to the critical question: Could this venture grow into a billion-dollar company?

The panel also discussed how AI startups can set themselves apart as competition intensifies. Bansal highlighted the necessity of specialized knowledge and a well-defined competitive approach. Neundorfer remarked that the companies that draw her interest are those that facilitate new behaviors instead of merely enhancing existing processes step-by-step.

Agarwal provided more pragmatic advice for founders, suggesting they clarify how AI technology powers their product; outline direct go-to-market approaches; and prove how their business will function more efficiently than established players.

It’s also essential to be transparent about the competitors in the field, she mentioned. “Some of you have lost a bit of credibility with me because you did not mention them on your slide,” she told the founders present.

In conclusion, the investors offered guidance on navigating the swiftly changing landscape. Agarwal encouraged founders to keep abreast of industry trends. Neundorfer advised maintaining connections within founder networks to exchange tools and insights.

Bansal’s advice was more straightforward: “Concentrate on developing your product.”

As social media becomes more dependent on algorithmic feeds, content creators are adapting to a new reality: Posting something doesn’t guarantee your audience will see it.

“I believe 2025 marked the year when algorithms completely dominated, rendering follower counts meaningless,” LTK CEO Amber Venz Box stated to TechCrunch.

This revelation isn’t new for creators — Patreon CEO Jack Conte has passionately highlighted this issue for years — but throughout the year, the broader industry has responded to this shift in various manners, spanning from influencers to streamers.

According to the executives TechCrunch interviewed regarding the impending future of the creator economy, content creators are discovering new methods to foster and nurture connections with their audiences — some serving as remedies to AI clutter, while others are inundating the space with a fresh kind of clutter themselves.

Box’s firm, LTK, links creators with brands via affiliate marketing, where creators earn commissions on recommended products. This business model wholly relies on audiences maintaining faith in individual creators. With worries about fragmentation in the creator-audience connection, this could pose a significant risk to the company.

However, a study commissioned by Northwestern University revealed that trust in creators rose by 21% year-over-year, which pleasantly surprised Box.

“If you had asked me at the start of 2025, ‘Will trust in creators increase or decrease?’ I would’ve likely said decrease, as people realize it’s a business — they comprehend how it functions,” she remarked. “However, AI has led people to shift their trust towards real individuals with genuine experiences.”

By this, Box implies that consumers are increasingly inclined to seek out content from the creators they recognize and trust. The study indicated that 97% of chief marketing officers plan to increase their influencer marketing budgets in the upcoming year.

Nonetheless, managing these relationships isn’t simple. LTK creators, dependent on affiliate earnings, are counting on this AI-induced skepticism to guide audiences toward more direct interactions through paid fan communities or less algorithm-centric platforms like LTK. For different types of creators, such as streamers, video podcasters, and short filmmakers, the approach to maintaining their audience often resembles growth hacking.

Clipping armies of teenagers

As Sean Atkins, CEO of the short-form video production company Dhar Mann Studios, expressed, “In a world dominated by AI and algorithms, where individuals trust other humans more amidst the micro division of attention, how do you market when control is elusive?”

Eric Wei, co-founder of Karat Financial, a financial service for creators, pointed out that creators possess a new advantage: teams of teenagers on Discord whom creators compensate for clipping their content, which those same teenagers then share extensively on algorithm-driven platforms.

“This has been happening for a while,” Wei explained. “Drake does it. Many of the largest creators and streamers engage in this — Kai Cenat [a leading Twitch streamer] participates — generating millions of impressions… If it’s algorithmically driven, clipping makes perfect sense, as it can originate from any random account with high-quality clips.”

Wei anticipates that clipping will gain even more traction this year as a response to the fragmentation in social media relationships. Even the most prominent creators face challenges in reaching their fans directly, which drives them toward clipping. While achieving virality on these algorithmic platforms is certainly easier with a substantial follower base, there’s no requirement for a prior record on a platform to determine whether a video should be widely shared. Thus, if these “clippers” share a brief highlight from specific creators’ streams, they can profit based on the views the video garners.

“Clipping seems like an evolution of meme accounts,” Glenn Ginsburg, president of QYOU Media, which creates content for younger audiences, conveyed to TechCrunch. “It has become a competition among creators to take this content and disseminate it broadly, almost racing to see who can amass the most views on the same intellectual property.”

Reed Duchscher — founding CEO of Night, the talent management agency representing Kai Cenat and other leading creators — expertly guides creators in maximizing their virality. As MrBeast’s former manager, Duchscher fostered the fast-paced, attention-grabbing style that transformed MrBeast from a YouTuber into an empire. He also oversees Kai Cenat’s clipping strategy, although Duchscher doesn’t share Wei’s enthusiasm regarding its wider potential.

“Clipping is crucial for a creator, as you need to saturate the space with content, and it’s a great method to increase visibility,” Duchscher stated to TechCrunch. “However, it’s also quite challenging to scale, as there are only a limited number of clippers online, making substantial media investments complicated.”

Perhaps clipping is effective now because the strategy has not yet gained such prevalence that it’s perceived as spam.

“The creator benefits by getting more of their content out,” Wei mentioned. “The clippers benefit because this group of teenagers is receiving payment. Everyone benefits, except that if this trend continues, we may end up with a surfeit of low-quality content.”

Niche appeal prevails

The rise of low-quality content on social media has become significant enough that Merriam-Webster designated slop as its word of the year.

“Over 94% of individuals claim social media is no longer social, and more than half are reallocating their time to smaller, niche communities that they know are genuine and that they can engage with,” Box noted, referencing platforms like Strava, LinkedIn, and Substack. 

As the connection between a creator and their audience grows increasingly challenging, Duchscher anticipates that creators with more defined niches will thrive — he believes that “macro creators” like MrBeast, PewDiePie, or Charli D’Amelio, who gather hundreds of millions of followers, will become even more difficult to replicate.

Citing success stories like Alix Earle or Outdoor Boys, who boast millions of followers but lack widespread appeal, Duchscher emphasizes, “Algorithms have become adept at presenting us with precisely the content we desire. It’s increasingly tough for a creator to break into every specific algorithm.” 

Atkins concurs, asserting that the creator economy encompasses much more than merely entertainment. “The creator economy is often viewed through the entertainment lens, which I consider a mistake, as contemplating the creator economy is somewhat akin to thinking about the internet or AI — it’s set to influence everything.”

Atkins cites the gardening creator brand Epic Gardening as an illustration. What began as a YouTube channel now holds a tangible, significant position in the gardening sector.

“Epic Gardening acquired the third-largest seed company in the United States, so now he’s the third-largest seed company [owner], as a content creator,” he remarked.

Although the creator economy is in a state of flux, it remains a robust industry — one well-versed in navigating the whims of the algorithm, persevering for decades, even if those unfamiliar may view it as an entirely new domain.

Creators are “literally influencing everything,” Atkins stated. “I bet there’s a content creator who specializes in cement mixing for skyscrapers.”

In the early months of 2025, the AI sector had no limits on funding. However, a shift in atmosphere began to emerge during the latter half of the year. 

OpenAI successfully secured $40 billion at a staggering $300 billion valuation. Safe Superintelligence and Thinking Machine Labs each garnered $2 billion in seed funding prior to launching any products. Even novice founders attracted investments on a scale previously reserved for major technology companies. 

These enormous investments were matched by equally impressive expenditures. Meta spent close to $15 billion to secure Scale AI CEO Alexandr Wang and invested many more millions to lure talent from competing AI labs. At the same time, the largest players in AI committed to nearly $1.3 trillion for future infrastructure development. 

The first half of 2025 retained the enthusiasm and investor interest seen in the previous year. Recently, though, this sentiment has shifted, prompting a kind of reality check. Extreme enthusiasm for AI and the associated lofty valuations persist, yet this optimistic outlook is now mingled with worries about a potential AI bubble, user safety, and the feasibility of sustaining technological advancements at such a rapid pace. 

The period of unreserved endorsement and celebration of AI is slightly receding. With it comes increased scrutiny and inquiries. Can AI firms maintain their growth rate? Does scalability in the post-DeepSeek landscape necessitate billions? Is there a viable business model that can generate a fraction of the multi-billion investments? 

We’ve observed every development closely. Our most read stories of 2025 reveal the underlying truth: an industry encountering a reality check while proclaiming its intent to alter reality itself. 

How the year commenced

The largest AI laboratories expanded significantly this year. 

In 2025 alone, OpenAI secured a SoftBank-led round of $40 billion at a post-money valuation of $300 billion. Reports suggest that the company has investors like Amazon interested through compute-related deals and is in negotiations to raise $100 billion at an $830 billion valuation. This would position OpenAI closer to the $1 trillion valuation it allegedly aims for in an upcoming IPO next year. 

Competitor Anthropic also raised $16.5 billion this year through two rounds; its latest funding round increased its valuation to $183 billion with backing from significant players like Iconiq Capital, Fidelity, and the Qatar Investment Authority. (CEO Dario Amodei revealed in a leaked memo that he felt “not excited” about taking funds from authoritarian Gulf states.) 

Then there’s Elon Musk’s xAI, which attracted at least $10 billion this year after taking over X.

Additionally, we’ve observed smaller, emerging startups gaining momentum from eager investors. 

Former OpenAI chief technologist Mira Murati’s venture, Thinking Machine Labs, landed a $2 billion seed round at a $12 billion valuation despite disclosing minimal details about its product lineup. Vibe-coding startup Lovable raised $200 million in a Series A funding round just eight months after its inception; this month, Lovable secured an additional $330 million at a nearly $7 billion post-money valuation. Not to be overlooked, AI recruiting startup Mercor raised $450 million this year across two funding rounds, the most recent elevating its valuation to $10 billion. 

Such extraordinarily high valuations continue to occur despite relatively low enterprise adoption rates and significant infrastructure challenges, fueling anxieties about a potential AI bubble. 

Build, baby, build

For larger companies, these figures don’t manifest out of thin air. To validate those valuations, massive infrastructure development is essential. 

This has resulted in a self-perpetuating cycle. Capital raised to support computing is becoming increasingly linked to agreements where those same funds are redirected back into chips, cloud services, and energy, as illustrated by OpenAI’s funding tied to Nvidia’s infrastructure. In practice, this obscures the line between investment and customer demand, raising concerns that the AI boom is being sustained by circular economics rather than genuine usage.

Some of the most significant deals this year fueling the infrastructure expansion included: 

• Stargate, a cooperative venture involving SoftBank, OpenAI, and Oracle, which allocates up to $500 billion for AI infrastructure in the U.S. 
• Alphabet’s purchase of energy and data center infrastructure provider Intersect for $4.75 billion, coinciding with the company’s announcement in October to increase its compute expenditure to $93 billion in 2026.
• Meta’s expedited expansion of data centers, elevating its projected capital expenditures to $72 billion in 2025 as the company strives to secure adequate compute resources for training and operating next-generation models. 

Nevertheless, signs of instability are beginning to surface. A financial partner, Blue Owl Capital, recently withdrew from a planned $10 billion Oracle data center deal associated with OpenAI capacity, highlighting the fragility of some financial arrangements. 

Whether all this expenditure will materialize remains uncertain. Grid limitations, skyrocketing construction and energy costs, along with mounting opposition from residents and policymakers — including calls from figures like Sen. Bernie Sanders for restrictions on data center growth — are already stalling projects in various areas. 

Even as AI investments remain substantial, the reality of infrastructure is starting to moderate the excitement. 

The expectation reset

During 2023 and 2024, each significant model release seemed groundbreaking, showcasing new capabilities and fresh motivations to embrace the hype. This year, however, the allure waned, epitomized by OpenAI’s GPT-5 launch. 

Although it carried weight on paper, it didn’t resonate like earlier releases such as GPT-4 and 4o. Similar trends were apparent across the sector, where advancements from LLM providers were less revolutionary and instead more gradual or domain-specific. 

Even Gemini 3, which excels in various benchmarks, was only significant in that it allowed Google to regain parity with OpenAI — leading to Sam Altman’s notorious “code red” memo and OpenAI’s struggle to retain its dominance.

This year also saw a shift in expectations regarding the emergence of frontier models. The launch of DeepSeek’s R1, its “reasoning” model that competed with OpenAI’s o1 on essential benchmarks, demonstrated that new labs could produce credible models quickly and at a fraction of the cost. 

From model breakthroughs to business models

As the magnitude of advances between new models narrows, investors are shifting their focus away from mere model capacity to what surrounds that core technology. The pressing question is: Who can transform AI into products that customers will depend on, financially commit to, and seamlessly integrate into their workflows? 

This transformation is manifesting in numerous ways as companies identify effective strategies and discover what consumers are prepared to accept. AI search startup Perplexity, for instance, briefly entertained the possibility of monitoring users’ online activities to deliver highly personalized advertisements. Concurrently, OpenAI was rumored to be contemplating charging up to $20,000 each month for specialized AI, indicative of how aggressively firms are probing what customers may be willing to spend.

Above all, attention has shifted to distribution. Perplexity aims to remain relevant by introducing its own Comet browser with autonomous capabilities and paying Snap $400 million to enhance search in Snapchat, effectively purchasing entry into established user pathways. 

OpenAI is pursuing a similar approach, working to evolve ChatGPT from a mere chatbot into a full-fledged platform. The organization has introduced its own Atlas browser and additional consumer-oriented features such as Pulse, while also appealing to enterprises and developers through app launches within ChatGPT itself. 

Google is leveraging its existing position. On the consumer front, Gemini is being integrated directly into products like Google Calendar, while on the enterprise side, the company is maintaining MCP connectors to solidify its ecosystem against challenges. 

In a landscape where distinguishing oneself solely through new models is increasingly challenging, controlling the customer relationship and business model emerges as the true competitive edge. 

The trust and safety vibe check

In 2025, AI companies faced unprecedented levels of scrutiny. Over 50 copyright lawsuits made their way through the judicial system, while reports of “AI psychosis” — the outcome of chatbots reinforcing false beliefs and allegedly leading to numerous suicides and other life-threatening situations — triggered demands for trust and safety reforms. 

Some copyright disputes concluded — such as Anthropic’s $1.5 billion settlement with authors — but most remain unresolved. The dialogue appears to be evolving from a resistance to using copyrighted material for training to requests for compensation. (Refer to: New York Times is suing Perplexity for copyright infringement.)

Concurrently, concerns regarding mental health in relation to AI chatbot interactions — and their overly flattering responses — emerged as a significant public health concern after multiple suicides and dangerous delusions in both teens and adults following extended chatbot engagements. The repercussions have included lawsuits, widespread alarm among mental health practitioners, and swift policies like California’s SB 243, which governs AI companion bots.

Most revealing are the calls for constraints not coming from typical anti-tech critics. 

Industry leaders have raised alarms regarding chatbots that amplify engagement, and even Sam Altman has cautioned against excessive emotional dependency on ChatGPT. 

Even the labs themselves began to express concern. Anthropic’s safety report from May captured Claude Opus 4 trying to blackmail engineers to avoid its own termination. The implication? Rapid scaling without comprehending what has been developed is no longer a sustainable approach.

Looking forward

If 2025 marked the inception of AI maturing and confronting difficult inquiries, 2026 will be when it must provide answers. The excitement cycle is beginning to diminish, compelling AI companies to validate their business models and demonstrate tangible economic value.

The age of “trust us, the returns will materialize” is approaching its conclusion. What lies ahead will either serve as a vindication or a reckoning that renders the dot-com crash a mere blip in trading for Nvidia. It’s time to make your predictions. 

A surge of AI voice recording devices such as Omi, Bee, and Friend has emerged, designed to capture your voice and enable interaction with an AI chatbot. Bee has been purchased by Amazon, and products like the Stream ring by Sandbar and a novel AI ring from previous Pebble founder Eric Migicovsky are anticipated to debut next year, yet the effectiveness of wearable AI gadgets remains uncertain.

In this landscape, Plaud is flourishing by appealing to professional users with a unique strategy: a credit card-sized recording unit that conveniently fits into your wallet. The company claims to have delivered over a million units, with more than half of its clientele transitioning to pro subscriptions.

The latest version, the Plaud Note Pro, was made available for preorder in August, two years after the initial Note, priced at $179. After a month of using the device, it has become a vital element of my daily essentials — its ultra-slim design facilitates this.

At merely 0.12 inches thick — roughly equivalent to three stacked credit cards — it stands as the slimmest AI recording device available and fits neatly in a wallet or can magnetically attach to the rear of your phone.

The company supplies a wallet-like case and a magnetic ring accessory for MagSafe-compatible phones, allowing you to affix the Note Pro to your iPhone or suitable Android device. Additionally, the device is exceedingly lightweight at 30 grams, so you won’t mind carrying the Note Pro in your wallet.

A significant distinction between Plaud and other AI wearables is that the Note Pro can record audio without needing to be linked to your phone. With 64GB of internal memory, the device can retain a substantial quantity of recordings without needing to transfer them to your phone or upload them to the cloud.

The Plaud Note Pro features four MEMS (Micro-Electro-Mechanical Systems) microphones to capture sound from all angles. While the company claims an effective audio range of 16.4 feet, I have successfully recorded discussions at conferences from a considerable distance away from the stage. The device also contains a voice processing unit for noise reduction, voice separation, and echo elimination.

This recording device boasts outstanding battery longevity. I attended a conference earlier this month with a fully charged unit and recorded several interviews and presentations while there. Subsequently, I utilized the device for phone call recordings and personal note-taking. Even after all that usage, the device retained 55% charge after 15 days. The company states that you can achieve 30 hours of continuous recording and 60 days of standby on a single charge.

Plaud’s latest gadget comes with a proprietary charger featuring a USB-C connector on the opposite end. It requires two hours to charge from 0%, preparing you for at least a couple of weeks unless you engage in extensive content recording.

One challenge associated with wearable AI gadgets is the necessity to visually confirm that the device is recording (or has ceased recording). Fortunately, the Plaud Note Pro is equipped with a small display indicating your recording status. You can also use a button during recording to emphasize a speaker’s point, which will be prominently displayed in the AI-generated summary. The screen additionally shows the remaining battery life.

There is a deliberate process involved in recording with this gadget. The device provides haptic feedback for starting and stopping recordings. The visual indicators and the action of pressing the button further assist in communicating to others in the meeting that you are recording the session.

You can simply opt to record sessions and send them to another AI transcription service you utilize. Plaud offers 300 minutes of free transcription each month. The company also allows you to personalize AI-crafted notes with templates tailored for various profiles and tasks. You also have the option to create your own template. The transcription is generally precise, and you can now access the recording, transcription, and notes through a designated website. The company has remedied the issue my previous colleague Brian Heater faced regarding tapping on a word and not hearing the corresponding recording.

While a pendant or pin-type design may be simpler to carry, the card-sized recorder provides superior microphones and more adaptable positioning options. If you frequently attend in-person meetings, investing in the $179 device is definitely worthwhile.

One of the most favorable remarks people have expressed regarding the electrical grid was … silence. The grid operates most effectively when it becomes inconspicuous. 

That understated position has shifted recently as wildfires in California and cold snaps in Texas have increased public awareness of the electrical grid. However, it was in 2025 that the electrical grid — along with concerns about demand, supply, pricing, and the pressure on natural resources — entered the limelight. A fresh wave of startups has appeared with a software-as-a-solution approach.

Electricity prices have surged by 13% in the U.S. this year due to an AI surge that has infiltrated unexpected areas, such as modifying supersonic jet engines for data center use and exploring methods to transmit solar energy from space.

Additionally, this growth trajectory is not anticipated to diminish; the energy consumption of data centers is expected to nearly triple over the next ten years. This projection has sparked consumer dissatisfaction over pricing and attracted the attention of environmental organizations that are demanding a nationwide halt on new initiatives. Utilities, traditionally working in the background, are now racing to enhance the grid and construct new power facilities capable of handling the demand — with the anxiety of a potential AI bubble burst always present.

This mixture of demand and apprehension might provide an advantage to software startups in the near future.

For instance, startups such as Gridcare and Yottar contend that there is already unused capacity on the grid and that software can assist in identifying it.

Gridcare has compiled information on transmission and distribution lines, fiber-optic networks, extreme weather patterns, and even community opinions to fine-tune the search for new sites and persuade utilities that the grid can accommodate more. The company claims to have already identified several neglected sites. Yottar locates areas with established capacity that coincides with the requirements of medium-sized users, facilitating their swift connection amidst the data center expansion.

Many other startups are utilizing software to connect extensive fleets of batteries distributed across the grid. These companies can transform these fleets into virtual power stations, supplying electricity to the grid during peak demand.

Base Power, for example, is establishing one in Texas by renting batteries to residents at relatively low rates. Homeowners can utilize the batteries for backup energy during outages, while Base can leverage them to avert outages by marketing the consolidated capacity to the grid. Terralayr is pursuing a similar path, although it does not retail batteries. Instead, Terralayr employs software to aggregate distributed storage resources already present on the German grid.

Other startups, such as Texture, Uplight, and Camus, are creating software solutions to integrate and synchronize diverse energy sources like wind, solar, and batteries. The aim is to ensure that by managing various resources, they will operate less idly and provide more support to the grid.

There’s also some optimism that software can aid in modernizing certain outdated aspects of the grid.

Nvidia, for instance, has teamed up with EPRI, an R&D organization in the power sector, to create industry-specific models in hopes of enhancing efficiency and resilience. Meanwhile, Google is collaborating with the grid operator PJM to leverage AI to navigate its backlog of connection applications from new electricity sources.

These transformations won’t occur instantaneously, but 2026 may mark the beginning of their implementation.

Utilities are often slow to embrace new technologies due to reliability concerns. They are also slow to invest in new infrastructure because of the expenses and longevity involved. Ratepayers and regulators have sometimes resisted when such projects threaten affordability.

However, software is more affordable, and if it can pass the reliability tests, the firms offering it stand a good chance of making significant headway.

And this could benefit more than just the startups promoting software. Ultimately, the grid requires some upgrading and expansion. Given the plethora of planned data centers and the electrification of vast areas of the economy, such as transportation and heating, additional power will be necessary. It would be unwise to overlook the potential of software in these scenarios. It’s cost-effective, adaptable, and quick to implement.