Editorial · Business & Funding
AI Investments Are Failing Because Companies Aren't Building the Right Culture
The race to adopt AI is on, with companies pouring billions into the technology. Yet, despite the hype, most are struggling to see real returns. According to a MIT study, 95% of generative AI pilots in large companies fail to produce measurable ROI. The issue isn't the technology itself but the flawed assumption that AI is a quick fix for inefficiency.
Many companies treat AI as a plug-and-play solution, layering advanced models over existing workflows. This approach accelerates outdated processes, leading to faster execution of bad strategies. Instead, successful AI adoption requires a fundamental shift in company culture-specifically, a culture of experimentation.
High-performing organizations don't just automate; they reinvent their workflows. They use AI not as a speed booster but as a tool for discovery. This means embracing the scientific method: formulating hypotheses, running experiments, analyzing data, and learning from results. For example, companies like Booking.com run over 25,000 experiments annually, fostering innovation through psychological safety, data-driven decisions, and a focus on learning.
The key to unlocking AI's potential lies in building this experimental culture. Leaders must prioritize curiosity over certainty, turning failure into a learning opportunity rather than a punishment. This shift transforms AI from a costly investment into a strategic asset, driving meaningful business outcomes.
Editorial perspective - synthesised analysis, not factual reporting.
If you liked this
More editorials.
The Quiet Shift That’s Transforming China’s Chip Industry - And It’s Bigger Than You Think
China’s chip industry is undergoing a quiet revolution, driven by unprecedented investment and innovation. While the global semiconductor market is projected to exceed $1.3 trillion in 2026-a growth milestone unmatched in decades-China is emerging as a key player in this explosive sector. The country’s major industrial firms saw profits surge 18.2% year-on-year in the first four months of 2026, with high-tech and chip-related sectors leading the charge. This growth isn’t just about numbers; it’s a testament to China’s strategic focus on becoming self-reliant in a sector critical to its AI ambitions. The U.S. has attempted to curb China’s advanced chip development through sanctions and export controls, but these measures have only accelerated domestic innovation. Chinese companies are increasingly adopting locally made chips to stabilize their supply chains, leading to a surge in integrated circuit exports. In the first 11 months of 2024 alone, China’s IC exports grew by an impressive 20.3% year-on-year, reaching $141 billion. This growth is no accident; it’s the result of robust policy support and clear government guidance, which have strengthened the industry’s foundation. China’s investment in its chip manufacturing capabilities has paid off. CITIC Securities predicts that shipments of domestic computing chips will at least double in 2026, driven by the explosive development of AI applications like Agent multimodal technology. This growth is particularly evident in mature process technologies, which are now meeting most application needs for industries like automotive and manufacturing. While advanced nodes remain a challenge due to U.S. restrictions, China’s progress in self-sufficiency is undeniable. The Chinese government has also doubled down on fostering innovation through initiatives like the National Integrated Circuit Industry Investment Fund. With its third phase launched in 2023, this fund has injected billions into the sector, ensuring that China remains on track to become a global leader in semiconductors. Companies like Huawei have already made strides in domestic chip production, with the Mate 70 series featuring 100% domestically produced chips-a clear signal of China’s commitment to reducing reliance on foreign technology. Looking ahead, China’s chip industry is poised for sustained growth, driven by both domestic demand and global opportunities. The U.S. may have hoped to slow down China’s progress, but the reality is that these restrictions have only motivated Chinese firms to innovate independently. As the race for AI dominance intensifies, China’s quiet shift in the chip industry is not just a response to external pressures-it’s a strategic move to secure its position at the forefront of this transformative technology. The future of semiconductors is being rewritten, and China is leading the charge.
The Subscription Fatigue Crisis: How AI Startups Are Surviving the Creator Economy Collapse
The creator economy is in turmoil. Creators - from professional influencers to weekend hobbyists - are drowning under a deluge of subscription services. Each month, they’re shelling out for text generators, video editors, image enhancers, and audio cleaners, all claiming to revolutionize their content creation process. But as the numbers reveal, this model is unsustainable. The global AI market may be skyrocketing, but consumer patience isn’t keeping pace. Recent data shows that if 1.8 billion AI users paid an average of $20 a month, it would generate $432 billion annually. Yet, reality paints a starkly different picture: only 9% of consumers pay for more than one AI subscription. This shift signals a growing resistance to the endless array of niche tools flooding the market. Creators are tired of juggling multiple apps - exporting files, managing logins, and switching tabs just to produce a short Reel. They want simplicity. Enter GlamAI. This startup is disrupting the status quo by offering a one-stop solution for creators. Instead of forcing users to choose between specialized tools, GlamAI aggregates premium foundational models under one interface. But it doesn’t stop there - it also boasts its own proprietary model, tailored specifically for visual editing and social media aesthetics. This dual-threat approach is proving effective: it gives creators access to familiar big names while hooking them with a specialized tool that understands the nuances of their work. GlamAI’s success is even more remarkable considering its bootstrapped status. With just 50 employees, no venture capital backing, and a focus on optimizing open-source models through smart patents, it has managed to carve out a niche in a competitive market. Its rise highlights a shift toward leaner operations and smarter tools that prioritize creator needs over feature bloat. The broader implications are clear: the era of scattered subscriptions is coming to an end. Creators are voting with their wallets - and the startups that listen will survive. The future lies in aggregation, specialization, and simplicity. For AI startups, the challenge is not just about building better tools but about understanding the growing impatience of their users. As the creator economy evolves, one thing remains certain: creators want to create, not manage a tech stack. Startups that can deliver on this promise will thrive - and those that don’t adapt will be left behind. The age of the "everything app" is here, and it’s not just about surviving - it’s about redefining how creators engage with AI tools in a way that respects their time and creativity.
The AI Race Heats Up: Applied Digital's $7.5 Billion Bet on Next-Gen Data Centers
Applied Digital is doubling down on its bet to become a leader in the rapidly evolving AI infrastructure market. With a 15-year lease agreement secured at its Delta Forge 1 campus, the company has locked in $7.5 billion in contracted value-a move that signals confidence in both its business model and the future of AI workloads. This deal, signed with a high-investment-grade hyperscaler, not only cements Applied Digital's position in the industry but also underscores the growing demand for purpose-built data centers designed to handle the intensive computational needs of artificial intelligence. The agreement at Delta Forge 1 is part of a broader strategy that has already seen the company secure over $23 billion in total contracted lease revenue. More than half of this figure now comes from investment-grade customers, a significant achievement that reduces financial uncertainty and positions Applied Digital for stable growth as it scales its operations. This level of customer commitment reflects the increasing importance of reliable, scalable AI infrastructure-both for hyperscalers looking to maintain competitive edges and for companies aiming to integrate advanced AI capabilities into their operations. Applied Digital's focus on high-density, energy-efficient facilities is a key differentiator in an industry grappling with sustainability and operational efficiency challenges. The company's proprietary waterless cooling technology exemplifies its commitment to innovation, allowing it to deploy data centers faster and more sustainably than traditional providers. This strategic advantage, combined with its ability to attract top-tier hyperscalers, suggests that Applied Digital is not just a player in the AI infrastructure game-it's setting the pace. Looking ahead, the company's plans for additional financing-up to $300 million in senior secured bridge financing and an additional $300 million revolving credit facility-indicate a clear intention to accelerate growth. These funds will support ongoing development at Polaris Forge 1 Building 3 and other projects across its platform. As Applied Digital continues to execute on its construction timelines, it is building a portfolio that not only meets current demand but also anticipates the future needs of AI-driven industries. The stakes are high for Applied Digital as it navigates this fast-moving landscape. While its strong pipeline and customer commitments provide a solid foundation, challenges such as managing debt levels and generating consistent free cash flow remain critical hurdles. The company's ability to overcome these obstacles will determine whether it can fully realize the potential of its $7.5 billion bet-and continue leading the charge in the AI race. In conclusion, Applied Digital's latest move signals a bold vision for the future of AI infrastructure. By securing long-term, high-value contracts and investing in cutting-edge technologies, the company is positioning itself as a key player in an industry poised for exponential growth. Whether it can sustain this momentum will depend on its execution, innovation, and ability to scale-factors that will be closely watched by both investors and hyperscalers alike.
The Hidden Cost of SoundHound AI's Growth: How Legal Scandals and Stock Issuances Are Undermining Its Value
SoundHound AI has been on a rollercoaster ride in the stock market, with its shares skyrocketing over the past few years. However, beneath the surface of its impressive revenue growth and acquisition spree lies a troubling reality that investors are starting to notice. The company's recent legal probes, shareholder lawsuits, and a $300 million at-the-market stock offering have created significant headwinds, casting doubt on its long-term viability. The legal troubles facing SoundHound AI are no small matter. Multiple investigations into governance, internal controls, and acquisition accounting are now underway, with plaintiffs filing lawsuits against the company. These issues not only divert management attention but also pose a serious threat to its financial health. As of June 7, 2026, shares of SoundHound AI (NasdaqGM:SOUN) have fallen by over 30% year-to-date, closing at $7.39-a stark contrast to the highs of over $22 seen just a few months prior. Adding to these challenges is the company's recent decision to raise $300 million through an at-the-market stock offering. While this move aims to bolster its balance sheet and support future acquisitions, it has raised concerns about potential dilution for existing shareholders. The timing of this offering could not be worse, coming on the heels of weak financial performance and increasing regulatory scrutiny. Despite these setbacks, SoundHound AI continues to push forward with ambitious plans. Its acquisition of LivePerson, a customer service messaging platform, is expected to close in the second half of 2026. This deal will give SoundHound access to a large customer base and expand its reach in the conversational AI space. However, integrating LivePerson's struggling business into its own operations could prove to be a significant challenge, especially given the company's already high cash burn rate. SoundHound AI's financials tell a story of rapid growth but also of underlying issues. While revenue surged by 52% year-over-year in Q1 2026, the company's gross margin declined sharply, and it reported a net loss of $25 million for the quarter. This is despite efforts to improve margins through moves to proprietary AI models. The worsening cash flow situation further complicates matters, with SoundHound burning through $26.3 million in operating cash flow during Q1 alone. Investors must also consider the broader context of the voice AI space. While it is a fast-growing industry, it is highly competitive, and simply acquiring more companies may not be a sustainable strategy for long-term success. The market has already punished SoundHound AI's stock for its struggles, but the company still faces significant risks. Looking ahead, SoundHound AI's ability to turn things around will depend on several factors. It needs to demonstrate that it can improve its financial performance while navigating legal challenges and integrating new acquisitions without overextending itself financially. If it fails to do so, the high costs of its growth strategy could ultimately undermine its value. In conclusion, SoundHound AI's recent troubles highlight the hidden costs of rapid expansion in a competitive industry. While the company has made strides in growing its revenue, its legal and financial challenges raise serious questions about its future. Investors must weigh these risks carefully before deciding whether to hold or sell their shares in this high-stakes game of growth versus stability.
The Quiet Shift in AI Spending: Why Companies Are Missing Out on Real Value
The AI revolution is here, but companies are pouring billions into the wrong places. While 93% of global AI budgets go toward technology-like models, chips, and software-just 7% is spent on training people to use these tools effectively. This spending imbalance isn’t just wasteful-it’s counterproductive. Research from Boston Consulting Group shows that 70% of AI value comes from people and processes, yet companies are neglecting this critical area. The result? Slow adoption, limited impact, and poor returns. Even though AI adoption has surged-from 55% in 2023 to 88% in 2025-only 39% of organizations see measurable benefits for their bottom line. This isn’t a technology problem; it’s a people problem. Companies are building faster than they’re training, leaving employees unprepared to harness AI effectively. Worse yet, trust in AI is eroding. Despite greater access to generative tools, workplace use has dropped by 15%. Employees are turning to “shadow AI” tools outside company policy, showing confusion and lack of support. Trust in generative AI fell by 38% between May and July 2025, highlighting a deeper issue: without the right skills, powerful tools don’t transform businesses-they create risk and distrust. The solution isn’t just technical training. While upskilling employees on how to use AI tools is important, it’s not enough. The real need is for cognitive capabilities-how people interpret, question, and apply AI outputs. These aren’t soft skills; they’re critical thinking skills that help employees recognize bias, challenge assumptions, and communicate effectively with AI. For example, bias awareness is crucial. Cognitive biases like confirmation bias amplify when AI generates polished, credible outputs that confirm existing beliefs. Training employees to spot these patterns can help them use AI as a tool, not a crutch. The goal isn’t to distrust AI but to use it mindfully, questioning its conclusions and evaluating its suggestions critically. The shift in AI spending-from speed to discipline-is only beginning. Companies that focus on sustainable AI adoption by investing in people will outpace those stuck in the experimentation phase. The future of AI isn’t just about technology; it’s about building a workforce that can thrive with it.