As artificial intelligence has captivated the public imagination in 2023 with the release of widely accessible generative AI tools like ChatGPT, tech giants have been racing to invest tens of billions of dollars into building out their AI capabilities and infrastructure. However, as these companies report quarterly earnings, Wall Street analysts are increasingly questioning whether these massive AI investments will ever generate adequate returns. Some are even sounding the alarm about a potential AI bubble that may be on the verge of bursting.

The core issue is that despite the hype and buzz, current AI systems have yet to demonstrate clear paths to profitability. Companies like Amazon, Intel, Google and Microsoft are spending huge sums on AI R&D and infrastructure but have little revenue growth to show for it so far. Amazon’s stock plummeted 9% after an earnings report showing high AI costs cutting into margins. Intel’s shares cratered 25% on news of cost-cutting measures to offset AI investments. Even Microsoft and Google, seen as AI leaders, are tempering expectations — execs have stated it may take 10–15 years for substantial returns on AI to materialise, a timeframe at odds with the short-term focus of public market investors.

A recent report from Goldman Sachs titled “Gen AI: Too Much Spend, Too Little Benefit?” encapsulates these growing concerns. Analysts point out that for all the excitement around generative AI’s potential to boost productivity and enable new applications, the underlying technologies are not new breakthroughs, but rather scaled up versions of deep learning techniques that have existed for years. The path to translating these AI systems into profitable products and services at scale remains unclear. There are also questions about whether AI can solve many complex, ambiguous real-world problems that aren’t amenable to statistical pattern matching on big datasets.

This isn’t the first “AI winter” the field has weathered. Previous cycles of hype and disillusionment occurred in the 1970s and late 1980s, when early AI systems failed to live up to grandiose promises. Some observers worry we may be headed for another such letdown. Venture capitalist Peter Thiel has called AI “the next dot-com bubble.” Researcher Gary Marcus cautions that unrealistic expectations around AI’s general intelligence risk a collapse in funding and interest when reality sets in.

However, there are key differences in the current AI boom compared to past hype cycles that may help the field power through this uncertainty and scepticism:

1) The current wave of generative AI systems, while not fundamentally novel, represent a meaningful step-change in capabilities compared to what existed before. Tools like ChatGPT are sufficiently powerful and adaptable to enable a wide range of potentially useful applications. Even if 90% of the hype doesn’t pan out, the remaining 10% could still be transformative.

2) AI is being deployed at a far greater scale by tech giants with huge computing resources and vast troves of training data. Billions of users are interacting with AI on a daily basis through search, social media, and productivity tools. This tighter feedback loop between development and deployment could accelerate progress.

3) Even sceptical observers concede that the opportunity cost of underinvesting in AI likely exceeds the downside risk of wasting some money on it. Companies that ignore AI could get left behind. The economic incentives to chase the upside are strong.

4) Much of the AI spending is on infrastructure and talent that has other applications beyond just AI. Investments in data centres, ML engineers and researchers may pay off even if specific AI products fail.

So, while an AI downturn of some sort seems inevitable as the gap between hype and reality widens, it’s unlikely to be as severe as past AI winters. The key for companies working on AI will be to focus on specific high-value applications rather than chasing general intelligence, be pragmatic about timelines, and stay disciplined about costs.

Some of the most promising areas for AI to generate real value in the near term include:

• AI-assisted software development: Using AI to help write, document and test code
• Business intelligence: Automating data analysis, insight generation and reporting
• Content creation: Augmenting human creativity in designing digital assets and marketing
• Customer service: Powering chatbots and virtual agents to handle queries
• Scientific research: Accelerating discovery by mining literature and running simulations

Making tangible progress in high-leverage domains like these will be key to sustaining belief in AI’s longer-term potential. The current moment of reckoning about AI’s limitations and true value proposition is ultimately a healthy regulator on runaway hype. Resetting expectations and focusing on substance is exactly what’s needed. The AI pioneers who are solving real problems will be the ones who endure and thrive on the other side.

Consider these strategic insights:

• Niche AI Applications: Rather than chasing broad AI capabilities, focus on developing targeted AI solutions for specific high-value niches relevant to your business or industry that can deliver clear ROI.
• AI-Assisted Efficiency: Look for opportunities to leverage AI to streamline operations and boost productivity in areas like marketing content creation, data analysis, customer support, and repetitive back-office tasks.
• Upskill with AI: Proactively build AI capabilities in your workforce through training programs, strategic hires of people with AI/ML expertise, and upskilling initiatives to avoid getting left behind.
• Realistic Expectations: Have an eyes-wide-open view of both the potential and limitations of current AI. Set pragmatic goals and timelines, stay agile, and view AI as an iterative journey vs an overnight transformation.
• Explore New AI-Enabled Business Models: Brainstorm opportunities for new products, services or business models that are uniquely enabled by AI, especially those that align to Australia’s economic strengths in areas like natural resources, agriculture, and services.

Signals from the future:

Emerging trends that are likely to drive changes to the way we live, work and do business.

• Ikea Drones Will Employ AI to Work in Tandem with Fulfillment Employees — Retail TouchPoints — Innovative integration of AI-powered drones at Ikea streamlines operations, improves accuracy, and empowers employees to focus on value-driven tasks.
• ARPA-H announces awards to develop novel technologies for precise tumor removal — ARPA-H — Revolutionary ARPA-H PSI program funding cutting-edge surgical technologies for precise tumor removal and improved surgical outcomes.

Deep strategy:

Longer form articles rich with insights:

• The Three Traps That Stymie Reinvention — MIT Sloan Management Review — Innovation through strategic partnerships can lead to transformative breakthroughs, highlighting the importance of resource allocation and risk assessment in organisational reinvention.
• The transformation imperative for midcap companies — McKinsey — Zak Gaibi, who leads McKinsey’s Transformation Practice in Europe, the Middle East, and Africa, is joined by Kedar Naik, a senior partner with deep expertise in growth transformations, and Mauricio Janauskas, the managing partner of McKinsey’s Santiago office and the leader of the Transformation Practice in Spanish-speaking Latin America.Sean Brown: Why do you believe that midsize companies need to undertake transformations?

Business at the point of impact:

Emerging issues and technology trends can change the way we work and do business.

• How our ‘No DMs’ Slack policy transformed workplace communication — Fast Company — Unlocking innovation through diversity and inclusion creates a competitive advantage for businesses in today’s global market.
• AI’s Next Frontier: Top Trends That Will Impact Business Growth — Forbes — Innovation — Leveraging AI for strategic workforce integration, purpose-focused solutions, streamlined operations, enhanced cybersecurity, and personalised experiences drives business growth and efficiency.
• Employees are Getting Comfortable with Use of AI and Soon AI is Going to Become Mainstream in Organisations, Gartner — Digital Information World — AI and DEX are becoming mainstream, transforming how employees work. Organisations must prioritise DEX to enhance productivity and foster talent growth.
• Temple & Webster’s $1 billion business plan and “openness to try new things” — Inside Retail Australia — Leveraging AI and strategic investments, Temple & Webster achieves record revenue and efficiency gains, positioning for billion-dollar valuation and international expansion.
• Best Practices For Leaders Integrating AI Into The Workplace — Forbes — Innovation — Strategic planning, clear objectives, and leadership guidance are key to successful AI integration in businesses, emphasising cost-effectiveness, training, and regulatory compliance.

‍