Adopting AI Within The Workplace
Discover how to adopt AI in the workplace, explore key startup trends, see what’s capturing investor attention—and get ahead of the next big thing.
Welcome to NEW ECONOMIES, rooted in my experiences working in venture capital. From inside the venture world, one thing is clear: we’re not just in a tech boom—we’re living through a full-blown economic rewrite. New platforms, new business models, and evolving public market strategies are reshaping everything.
Join us on this journey—subscribe to NEW ECONOMIES to stay ahead of the curve in technology.
AI REPORT 2025
NEW ECONOMIES is dedicated to bringing you the latest technology trends that are reshaping our world. As AI continues to advance at an extraordinary pace, we’re thrilled to launch this year’s AI Report, which explores:
The current state of AI: Now and in the future.
How to adopt AI within the workplace across functions.
The latest platforms you should be using.
The startup fundraising landscape: The most active investors, the mega funding rounds and the latest fundraising activity.
The biggest challenges that lie ahead: What to expect.
The impact of AI: Where do we go from here?
Who will win: China or the U.S.?
AI Trends: Which trends operators and investors are most excited about.
Resources: Learn how to build the next big AI company. Read the pitch decks from companies who have raised billions.
Plus so much more…
This report would not have been possible without the support of our three incredible partners: MarkeTeam, Strawberry and Syncly.
A few words about them:
MarkeTeam
MarkeTeam.ai has developed the world’s first Large Language Model (LLM) purpose-built for marketing—powering a suite of autonomous AI agents that collaborate seamlessly across key functions like brand strategy, performance marketing, social media, content creation, and SEO.
Strawberry
Strawberry is a browser that comes pre-packed with your own personal AI companions. They can research across 100s of tabs in minutes, helping you collect structured data, create reports, cross-reference data sources and more. They can also interact with your tabs using their own mouse and keyboard where you can teach them to automate mundane work so that you can focus on what you do best.
Syncly
Syncly is a customer feedback intelligence platform built to help brands make sense of feedback from social, e-commerce, and CX channels. Syncly’s AI cuts through the noise of scattered, unstructured feedback from different sources to deliver a clear, comprehensive view of what customers are really saying, delivering actionable insights that power better customer experience and smarter brand decisions.
Let’s explore what is happening across the current landscape.
THE STATE OF AI NOW AND IN THE FUTURE
AI continues to accelerate at a rapid pace, with breakthroughs happening across foundational models, multimodal capabilities, and agentic workflows. Tools such as ChatGPT-4, Claude, and Gemini have all made AI more accessible for everyone.
Let’s not forget, this has all happened in just a few years. Before ChatGPT launched in November 2022, AI was not as hot as it is today.
The broader conversation is shifting from “what can AI do?” to “how do we use it responsibly, efficiently, and at scale, and now?” As a result, ethical considerations, regulation, and long-term alignment have taken center stage, even as technical capabilities continue to improve.
In the startup world, AI is still the hottest vertical. From vertical SaaS tools to infrastructure platforms, founders are exploring ways to infuse intelligence into every layer of the stack. Venture capital is flowing into AI-native companies, though with more scrutiny than the 2023 hype wave such as, how will these startups generate revenue and when?
To give you an inside look into the startup world—Startups are experimenting with everything from AI agents that handle operations and research, to copilots for legal, healthcare, and engineering tasks. The competition is definitely heating up, especially in crowded categories like chat interfaces and workflow automation, but there’s still plenty of white space—especially in enterprise tools and domain-specific applications.
Inside the workplace, adoption of AI is becoming more strategic. Companies are moving beyond simple chatbot integrations and beginning to deploy AI for real productivity gains—think summarizing meetings, generating code, automating customer service, and optimizing internal knowledge retrieval. Companies are probably also, starting to think how can they use AI effectively to free up cash flow (i.e., reduce the size of their organization).
That said, adoption isn’t without its challenges. Many companies face roadblocks around integration, employee training, and maintaining trust in AI-generated output. There's also the ongoing tension between using general-purpose tools versus building custom solutions. Still, the trajectory is clear: AI is shifting from experimental to essential. The organizations investing in smart, thoughtful implementation now are positioning themselves to lead in a future where intelligent systems aren’t a bonus—they’re the baseline.
Let’s take a closer look at how far AI has already progressed in just the first five months of 2025.
AI’S PROGRESS IN 2025 SO FAR
AI is off to a strong start in 2025 and already in just the first five months, we are witnessing the category changing at incredible speed. A few key developments which have already changed this year:
DeepSeek
DeepSeek released DeepSeek-R1, a large language model utilizing chain-of-thought reasoning, achieving performance similar to OpenAI's o1 across various tasks.
Baidu's Ernie X1 and Ernie 4.5
Baidu introduced Ernie X1, an AI model comparable to DeepSeek R1 but more cost-effective. They also unveiled Ernie 4.5, an updated foundational model claimed to surpass OpenAI's GPT-4.5 in performance.
xAI's Grok-3
Elon Musk's xAI launched Grok-3, trained with significantly more computing power than its predecessor, Grok-2. Grok-3 is reported to outperform OpenAI’s GPT-4o on several benchmarks.
Google's Collaboration with MediaTek on AI Chips
Google is set to partner with Taiwan's MediaTek to develop the next generation of its Tensor Processing Units (TPUs). This collaboration aims to reduce production costs and reliance on current partners, with production scheduled for the following year.
Stability AI’s new AI model turns photos into 3D scenes
Stability AI has released a new AI model, Stable Virtual Camera, that the company claims can transform 2D images into “immersive” videos with realistic depth and perspective.
AI IPOs
We may see some of the first AI IPOs in 2025, with companies like CoreWeave and Scale AI potentially going public.
Launch of The Stargate Project
On January 21, a consortium comprising OpenAI, Oracle, SoftBank, and MGX announced The Stargate Project. This initiative aims to build a comprehensive AI infrastructure system in collaboration with the U.S. government, with an estimated investment of $500 billion over the next four years.
A $50M Fund
Perplexity AI launched a $50 million venture fund to back early-stage startups.
Nvidia announces two ‘personal AI supercomputers’
Nvidia at GTC 2025 announced a new lineup of “AI personal supercomputers” powered by the company’s Grace Blackwell chip platform.
Saudi Arabia launches Humain
Saudi Arabia’s Crown Prince, Mohammed bin Salman, launched an AI company called Humain to bolster the country’s artificial intelligence sector. The company plans to build out the country’s AI infrastructure, including the construction of data centers.
So, how far has AI evolved?
HOW AI HAS EVOLVED
Over the past five years, AI has changed faster than probably any other category, largely driven by breakthroughs in deep learning and the rise of large language models. Tools such as OpenAI's GPT-3 and GPT-4, Google’s PaLM, and Meta’s LLaMA models have showcased how machines can now understand and generate human-like language with remarkable fluency.
These models which are trained on massive datasets, can summarize articles, write code, compose music, and even engage in nuanced conversations. This leap in natural language processing has made AI more accessible and useful to everyday users and businesses.
Another major leap has been the rise of multimodal AI—systems that can process and generate not just text, but also images, audio, and video. Platforms like DALL·E, Midjourney, and Sora are enabling users to create art, design, and even short films using simple prompts. This shift has opened the door for AI to play a larger role in creative industries, from marketing to game development as we covered in our recently published annual creator economy report.
At the same time, concerns about ethics, bias, and regulation have moved to the forefront. As AI systems become more powerful, there's a growing focus on responsible AI development—ensuring transparency, fairness, and accountability with Governments and companies alike introducing AI policies.
In short though, the last few years have marked a shift from AI being a niche tech topic to a mainstream force shaping how we live, work, and interact with the world—forever.
Our market maps below illustrate the rapid change of the AI landscape since 2022. In just a few short years, we’ve seen a dramatic shift—from the early days of Generative AI to the emergence of advanced AI Agents. These snapshots highlight just how quickly the field is advancing.
We take a closer look at what is happening across the startup landscape.
THE STARTUP FUNDRAISING LANDSCAPE
Startups have been raising capital at record speed, with term sheets issued within days. Since Q1 2023, over $220 billion has been invested in AI startups—nearly $60 billion of that in just the first quarter of this year.
Each quarter has received the following investment amount into AI startups:
Q1 '23 – $16.5B
Q2 '23 – $12.1B
Q3 '23 – $15.0B
Q4 '23 – $13.3B
Q1 '24 – $13.9B
Q2 '24 – $24.8B
Q3 '24 – $22.0B
Q4 '24 – $44.0B
Q1 '25 – $59.6B
UPGRADE YOUR NEW ECONOMIES SUBSCRIPTION AND BECOME A PAID SUBSCRIBER TO ACCESS OUR ARCHIVE WHICH INCLUDES THOUSANDS OF AI STARTUPS.
THE 10 MEGA AI FUNDING ROUNDS IN RECENT YEARS
Databricks – $10B
In December 2024, Databricks raised $10 billion at a $62 billion valuation. The round was led by Thrive Capital, with participation from Andreessen Horowitz, Iconiq Growth, and others.
xAI – $6B
Elon Musk’s AI startup, xAI, secured $6 billion in May 2024, valuing the company at $24 billion. Investors included Valor Equity Partners, Andreessen Horowitz, Sequoia Capital, and Fidelity Management & Research Co.
OpenAI – $46B
In October 2024, OpenAI raised $6.6 billion, bringing its valuation to $157 billion. The funding round included investments from Thrive Capital, Microsoft, Nvidia, SoftBank, and others. Fast forward six months to March 2025, OpenAI received another cash injection of $40 billion led by SoftBank.
Anthropic – $4B
Anthropic, the developer of the AI assistant Claude, received a $4 billion investment from Amazon in November 2024, as part of a broader strategic partnership. They also received another cash injection of $3.5 billion lead by Lightspeed in March of this year.
Waymo – $5.6B
In October 2024, Alphabet invested $5.6 billion into Waymo, its autonomous vehicle subsidiary, marking the startup’s first raise since a $2.5 billion round in 2021.
Scale AI – $1B
Scale AI raised $1 billion in May 2024, led by Accel, with participation from Nvidia, Meta, and Amazon, valuing the company at $13.8 billion.
Figure AI – $675M
In February 2024, Figure AI secured $675 million in funding from investors including Jeff Bezos, Microsoft, Nvidia, Intel, and OpenAI, valuing the company at $2.6 billion.
Mistral AI – $645M
Mistral AI, a French AI startup, raised €600 million ($645 million) in June 2024, elevating its valuation to €5.8 billion ($6.2 billion). The round was led by General Catalyst, with contributions from existing investors.
Moonshot AI – $1B
In February 2024, Chinese AI company Moonshot AI raised $1 billion in a funding round led by Alibaba Group, bringing its valuation to $2.5 billion.
Wayve – $1.05B
UK-based AI startup Wayve secured $1.05 billion in May 2024, led by SoftBank, with participation from Nvidia and Microsoft. This Series C funding round is the largest for a UK AI-focused startup.
HOW TO ADOPT AI WITHIN THE WORKPLACE
With so much happening in AI, now is the time to adopt these tools in your company—and we're here to show you how.
To help you navigate, we have categorized the platforms into the following categories which include:
Build your own agent
Customer experience
Compliance
Data & Analytics
Design Tools
Dev Tools
Digital Workers
Finance
HR
Legal
Marketing
Productivity
Research
Sales
Security
DISCOVER THE LATEST TOOLS AND HUNDREDS OF OTHERS WHEN YOU UPGRADE AND BECOME A PAID SUBSCRIBER. YOU CAN ALSO GIFT A SUBSCRIPTION TO A FRIEND AT NO EXTRA COST. IT’S A NO-BRAINER.
HOW YOU CAN ADOPT AI ACROSS THE WORKPLACE
Adopting AI in the workplace begins with identifying repetitive, time-consuming tasks that can be automated to improve efficiency. From scheduling meetings with smart assistants to using AI-driven tools for data entry, transcription, or customer support, these small changes can significantly boost productivity. For example, companies are increasingly integrating AI-powered chatbots to handle basic inquiries, allowing human staff to focus on more complex issues that require emotional intelligence or critical thinking.
Beyond automation, AI can also enhance decision-making across departments. Marketing teams use AI to analyze customer behavior and optimize campaigns, while HR departments deploy it to screen resumes or predict employee retention trends. In operations, predictive analytics can streamline supply chains and flag potential disruptions before they happen. These capabilities not only save time but also lead to more informed, data-backed strategies that can improve overall business outcomes.
However, successful adoption requires more than just plugging in new tools—it involves a cultural shift. Organizations need to train employees on how to work alongside AI and foster a mindset of continuous learning. Transparency around how AI systems work and how data is used is essential to build trust. When implemented thoughtfully, AI doesn’t replace all workers—it empowers them, unlocking creativity, improving collaboration, and driving innovation across the workplace.
Companies are starting to see the impact. More AI. Less team members. But what does this mean for AI more widely and how can teams actually go around adopting AI? We take a closer look.
EXAMPLE: HOW TO ADOPT MARKETING AGENTS
Marketing teams are rapidly embracing AI technologies to enhance their capabilities and streamline operations. Currently, 68% of marketers use AI in their daily work. This adoption is driven by the need for increased productivity, continuous learning, and more informed decision-making.
In fact, 82% of marketers and communicators expect AI adoption to further boost productivity by applying AI across a variety of tasks, including data analysis, customer segmentation, personalized content creation, and social media management.
In the rapidly evolving AI marketing landscape, MarkeTeam.ai delivers a truly AI-native, full-suite solution purpose-built for modern marketing teams. Its proprietary models are engineered to drive exceptional performance across the full spectrum of marketing activities—from deep research, brainstorming, and strategic planning to creative development, multichannel execution, and ongoing optimization.
Unlike prompt-based tools, MarkeTeam deploys fully autonomous agents that proactively initiate tasks, make strategic decisions, and manage complete marketing work units from start to finish. These agents not only operate independently but also collaborate as a tightly orchestrated multi-agent system—each with specialized roles, shared context, and goal alignment.
This architecture enables advanced planning, adaptive strategy shifts, and compound reasoning across time and tasks—capabilities that are not achievable through conventional single-model approaches or instruction-tuned tools. They are also trained for advanced marketing tool use, ensuring seamless execution across live marketing environments.
Natively embedded in a brand’s assets, channels, and tech stack, the agents continuously learn and adapt—absorbing brand identity, monitoring market trends, tracking competitors, and responding in real time, 24/7.
This level of autonomy not only reduces the burden on human team members but also ensures consistent, high-quality output regardless of a user’s familiarity with AI systems. By deploying MarkeTeam, marketing teams unlock always-on intelligence and execution capabilities, driving higher output, lower acquisition costs, faster speed to market, and scalable growth with enterprise-grade performance at a fraction of the cost.
Examples of how you can adopt AI within the workplace for your marketing team by using the likes of MarkeTeam.
HOW TO ADOPT AI FOR CX TEAMS WITH SYNCLY
Syncly is a customer feedback intelligence platform that helps brands make sense of what customers are really saying—across social, e-commerce, and CX channels. By cutting through the noise of scattered, unstructured feedback, Syncly’s AI delivers a clear, actionable view that drives better customer experiences and sharper brand decisions.
More CX teams are turning to AI to elevate how they work. AI enables personalized customer interactions by analyzing sentiment in real time and resolving issues proactively. It automates repetitive tasks, freeing teams to focus on what matters. And by aggregating feedback from surveys, reviews, and social media, it surfaces trends and opportunities faster than ever. With tools like Syncly, which bring all of these capabilities together, streamlining this process is easier—and more effective—than building them yourself.
For CX teams, the case for AI is clear. It streamlines workflows, eliminates manual processes, and brings visibility to what used to be scattered and siloed. With the volume of customer feedback growing—and the stakes higher than ever—AI is no longer optional. It’s essential for scaling insights, spotting gaps, and making smarter, faster decisions.
Adopting AI starts with gathering the right data—from internal channels to public platforms like social or e-commerce. Once collected, AI assists in sorting and structuring the feedback using tailored taxonomies.
Sentiment analysis follows, helping teams understand how customers perceive the brand. Finally, AI turns this data into insights—surfacing patterns, powering dashboards, and ensuring everyone across the org is aligned around what customers are really saying.
Syncly handles the heavy lifting—from data collection to insight generation—making it easier for CX teams to integrate AI into their daily workflows and maximize business impact.
HOW TO ADOPT PRODUCTIVITY TOOLS BY USING INTELLIGENT BROWSERS SUCH AS STRAWBERRY
While AI has become increasingly accessible through specialized tools and platforms, many organizations struggle to integrate it into their existing workflows. Teams are often required to context-switch between standalone AI applications and their daily work environment. For professionals who spend most of their time in-browser, this means constantly toggling between AI assistants like ChatGPT and their actual work, with AI lacking crucial context about their tasks and tools, including LinkedIn, CRMs, communication channels, or analytics tools.
One solution is to adopt an AI native browser such as Strawberry, which automatically brings context aware AI into all your tabs. It helps you extract structured data from websites, cross-reference data across tabs, populate spreadsheets and documents, and even research for you by opening new tabs autonomously.
Individuals and teams use Strawberry to automate their mundane work so that they can focus on high value tasks. By bringing AI into a browser environment, Strawberry has become the first AI application that can actually use all the services you use on a daily basis, from LinkedIn and G-Suite to CRMs.
That’s where another strength of AI-native browsers comes in—they can act on your behalf across the web. Thanks to the rapid improvement of foundation models, browsers like Strawberry are able to complete entire tasks in your tabs using an AI controlled mouse and keyboard, all you need to do is ask. You can even leave the tab where your agent is working, and come back later to check in on its progress. These AI browsing capabilities open up a whole new world of automation possibilities. Any task you perform in your browser can now be delegated to your own personal AI agent that has access to the tabs you do.
Strawberry can open tabs for you in the background and do research on your behalf. Individuals and teams save the most time with Strawberry when automating research tasks like finding leads on LinkedIn, creating competitor analyses, extracting structured data at scale and populating spreadsheets automatically.
THE BIGGEST CHALLENGES
While AI is making a huge impact on all of us, the path forward is filled with complex challenges that organizations and governments must address. One of the most pressing issues is AI regulation and policy.
As AI systems become more capable and embedded in daily life, there’s growing concern over how they should be governed. From ensuring transparency in decision-making to protecting user privacy and preventing bias, regulators are racing to catch up with a technology that evolves faster than legislation. The stakes are high: without thoughtful regulation, we risk reinforcing systemic inequalities or creating systems that lack accountability.
Security is another major challenge. As companies increasingly rely on AI for critical operations, these systems become prime targets for cyberattacks. Adversarial attacks—where inputs are manipulated to fool AI systems—pose unique risks, especially in areas like finance, healthcare, and autonomous vehicles. On top of that, there's growing concern over data security.
A newer, emerging challenge is the rise of AI agents talking to one another—networks of autonomous systems capable of coordinating actions without direct human oversight. While this opens the door to unprecedented levels of efficiency, it also raises difficult questions about control, unintended consequences, and how these agents might interact in unpredictable ways. Ensuring alignment between agents, and between agents and human goals, is a frontier problem that’s still unsolved.
Other critical challenges include job displacement and workforce re-skilling, as AI begins to handle more cognitive tasks traditionally done by humans. While AI will create new roles and categories of work, the transition could be rocky for millions of workers if companies and governments don’t invest in education and up-skilling programs. There's also the question of model interpretability—how do we understand why an AI made a specific decision, especially in high-stakes environments like healthcare or criminal justice?
In short, the challenges are real—but they’re not insurmountable. Navigating this next chapter will require collaboration across sectors, bold leadership, and a commitment to building AI that is not only powerful, but also trustworthy and aligned with human values.
THE IMPACT OF AI
The impact of AI within the workplace is already transforming how organizations operate at every level.
From automating routine administrative tasks to enhancing decision-making with real-time data insights, AI is driving a shift from manual workflows to intelligent systems. Tools like AI-powered chatbots, scheduling assistants, and document summarizers are giving hundreds of hours back to their team so human-employees can do what they are good at—growing the business and focusing on generating revenue.
AI is also fundamentally reshaping the structure of teams and how work gets done. Departments like HR, finance, marketing, and customer support are now increasingly leaning on AI to boost productivity and improve outcomes.
For example, HR teams use AI to screen resumes and predict employee retention risks, while marketing teams leverage predictive analytics to personalize campaigns. In engineering and product development, AI assists with coding, debugging, and even brainstorming new features. This is leading to leaner teams that can do more with less, often achieving what previously required significantly larger headcounts.
However, the rise of AI also brings new challenges. Companies must navigate data privacy, workforce re-skilling, and questions of trust around AI-generated output. Employees need to understand not only how to use AI tools, but how to collaborate with them effectively. Meanwhile, leaders must think strategically about AI implementation—balancing efficiency with transparency and human oversight.
Long term, the organizations that will thrive are those that integrate AI not just as a tool, but as a cultural mindset. AI isn't just about automation—it's about augmentation. It empowers employees to work smarter, make better decisions faster, and unlock entirely new ways of solving problems. The workplace of the future is not man versus machine—it’s a partnership.
WHO WILL WIN? CHINA OR U.S.?
The race for AI dominance between China and the United States is shaping up to be one of the defining global competitions of the decade. Both nations bring distinct advantages to the table.
The U.S. leads in foundational research, cutting-edge innovation, and homegrown talent—thanks to its world-class universities, deep venture capital networks, and tech giants like OpenAI, Google DeepMind, Microsoft, and NVIDIA. The country’s open ecosystem encourages rapid experimentation and entrepreneurship, helping AI startups scale quickly and reach global markets.
China, on the other hand, brings unmatched scale, state-driven strategy, and a vast pool of data. With strong support from the government—through its national AI plan and direct funding—China is focused on becoming the world leader in AI by 2030.
Companies like Baidu, Tencent, Alibaba, and SenseTime are investing heavily in AI applications, particularly in computer vision, surveillance tech, logistics, and smart cities. Additionally, China’s more centralized control over data and regulation allows for faster deployment and integration of AI across public infrastructure.
The U.S. still holds a lead in developing frontier AI models and publishing high-impact research, but China is closing the gap rapidly by localizing capabilities, investing in semiconductor self-sufficiency, and building massive talent pipelines. The outcome of the race may ultimately depend not just on who builds the best models, but on how each country balances ethics, regulation, innovation, and impact at scale.
Based on our data, the U.S. is currently producing more AI technology unicorns. As of now, 61% of unicorns are based in the U.S. and 17% are based in China.
AI UNICORNS LANDSCAPE
Discover the latest 140+ AI unicorns, who they are and what they are building.
UPGRADE YOUR NEW ECONOMIES SUBSCRIPTION TO ACCESS ALL AI UNICORNS. IT’S A NO-BRAINER.
WHERE IS AI HEADING NEXT
The big question on many minds is: where is AI going next? While the technology has already reshaped how we work, communicate, and innovate, we’re still in the early innings of a much larger transformation. What’s clear is that AI is here to stay—and its rapid advancements are accelerating everyday—if you are not building in this category, you are potentially missing out on the greatest technology change of this generation.
The next wave won’t just be about smarter tools, but about autonomous systems that work together, learn independently, and drive outcomes with minimal human intervention.
One of the most exciting frontiers is the rise of AI agents that collaborate with one another. These agents—software entities with the ability to act, reason, and make decisions—are beginning to form multi-agent ecosystems where they can coordinate tasks, delegate responsibilities, and adapt dynamically to changes in real time. Imagine a product team where each phase of the design-build-launch cycle is handled by specialized agents, communicating and adjusting as conditions change. This shift could redefine productivity across nearly every sector.
But with all this potential comes a reality check: many AI startups won’t survive the next few years. The market is crowded, hype is high, and only those who can solve real problems, scale efficiently, and deliver measurable value will break through. As the ecosystem matures, we’ll likely see a thinning of the herd—where only the most adaptive and capital-efficient startups remain. This is a natural part of any innovation wave, and the AI space is no different.
Interestingly, we’re also seeing a rise in ultra-lean AI companies, where tiny teams—sometimes even solo founders—can build world-changing products. Thanks to powerful foundation models, APIs, and plug-and-play infrastructure, a team of two or three can now compete with what used to take dozens or hundreds of employees. This trend could usher in a new era of micro-startups, capable of scaling fast, staying nimble, and pushing the boundaries of what’s possible with minimal overhead.
Looking ahead, the question isn’t whether AI will continue to evolve—but how quickly and in what direction. Will we see a consolidation of tools under a few major platforms, or a flourishing of niche, specialized agents? Will regulation slow things down, or force smarter innovation? Regardless, the future of AI is going to be defined by the teams, tools, and ideas that prioritize adaptability, trust, and impact.
THE LATEST TRENDS TO WATCH
UI Design is changing…
UI design standards on the internet have been built on the premise that data is structured and relatively static. Today’s foundation models turns that premise on its head, and I’m excited to see how user interfaces evolve given the hyper dynamic nature of LLMs. What happens to the UI if the services you use have deep context about you, your preferences and what you’re working on? Will every website generate on the fly to be tailor-made just for you?
—Sebastian Thunman, Co-Founder at Strawberry
Decoding complex data…
In 2025, I’m excited to see how AI will continue to evolve and unlock its full strength by decoding complex, unstructured data—a challenge many solutions still face. While much of the industry focuses on content generation, the true competitive edge lies in extracting deep, quality, and actionable insights. Tools capable of delivering precise, granular analysis and surface meaningful patterns will transform decision-making and fuel growth in today’s data-driven world.
—Joseph Lee, Co-Founder and CEO at Syncly (YC W23)
AI Agents to be adopted within the workplace…
2025 will mark the year businesses stop thinking of AI as tools and start treating them as co-workers. This isn't just semantic—it's a fundamental shift in how organizations operate. While this transformation raises hard questions about the future of work, it's unlocking unprecedented opportunities for businesses that have been held back by resource constraints.
—Naama Manova-Twito, Co-Founder at MarkeTeam
The shift to Agentic employee companies…
In 2025, I am excited about the shift to Agentic employee companies—i.e. natively from the ground up where the employee base is more agentic than human. For example, an agentic law firm would respond in minutes, not days.
—Chris Tottman, Partner at Notion Capital
The super charger for non technical people…
In 2025, I am excited to observe how AI is democratizing technology access. Advancements such as low-code and no-code platforms, accessible AI libraries, and cloud-based services have made it remarkably easier for small teams—even non-programmers—to quickly build products. This surge in empowerment is driving innovation forward, and I’m excited to see where it leads throughout 2025.
—Lorenzo Mengolini, Lead Product Designer at Revolut
The impact of tone of voice…
In 2025, I am particularly excited about the opportunities that AI combined with human creativity will open when it comes to content strategy. Whether it comes to SEO, content personalization in campaigns, CRM, tutorials and Social Media, AI will help creating tons of topical, fresh, localized and personalized content. What will be make-or-break for the brands however, is how the brand tone of voice and personality will shine through and help emerge and create differentiation. This can only be done through human expertise and experience.
—Flore Tyberghein, Chief Marketing Officer at BlaBla Car
The ROI on AI will peak…
In 2025, AI ROI (return on investment) will peak as compute costs will plummet as real business impact is realized. Open source models will lead to much lower cost use-case specific models deployed on commodity hardware. At the same time, experimental use cases from 2024 will evolve into full production platforms as AI is made more reliable.
—Cy Khormaee, Vice President, Product Management at Attentive
Voice becoming our primary interface…
In 2025, I am excited about voice AI becoming our primary interface with technology. After seeing my aunt chat naturally in Serbian with AI, I'm convinced it will bring AI to the next billion users.
—John Milinovich, Head of GenAI Product at Canva
Copilots for all jobs…
In 2025, there will be copilot for all jobs, not just knowledge workers. There will be copilots with hardware+software+AI+machine vision solutions for general contractors, architects, and policeman etc.
—Sri Batchu, Chief Marketing Officer at The RealReal
Using AI thoughtfully…
In 2025, I’m excited about how AI is challenging us to work smarter, not just faster. It’s not just about automating processes but using AI thoughtfully—to better understand users and create work that truly resonates. The power of AI lies in how we use it: not to produce more for the sake of it, but to deliver work that is impactful, meaningful, and adds real value.
—Lihi Lotker, Former Head of Marketing at Coho AI
Vertical specific applications…
In 2025, I am most excited about the convergence of Generative AI and vertical-specific applications. In 2025, we’re seeing AI evolve beyond generic tools into specialized solutions—whether it’s automating legal processes, creating personalized marketing strategies, or enhancing digital health through AI-driven diagnostics and preventative care. This shift allows businesses to address real-world problems with precision and scale.Lastly, the rise of AI-augmented decision-making to combat information overload is transformative. As AI becomes a horizontal layer across industries, it’s enabling faster, smarter choices, which will drive innovation in everything from financial inclusion to climate tech.
—Azmat Ali, Former Head of Marketing, Rider Products at Lyft
AI agents becoming indispensable for software developers…
In 2025, I am excited about AI agents becoming indispensable for software developers. These intelligent assistants will not only automate mundane tasks like code generation and debugging but also act as true collaborators.
They will understand the developer's intent, suggest optimal solutions, and even help refactor code for better maintainability. This shift will not only dramatically increase developer productivity but also empower developers to focus on higher-level creative tasks and innovation.
—Nandan Thor, Head of AI at Palo Alto Networks
Advancements in quantum computing…
In 2025, my wishful thinking is to hear less about GenAI and focus more broadly on machine learning applications. I’m especially curious to see how machine learning as a service evolves and would love to start hearing more news about advancements in quantum computing. Could AI and quantum computing become the perfect combo?
—Stefania Bonà, Head of AI Products at Trustly
AI co-workers operating on autopilot…
In 2025, I am excited about the transformative potential of AI-powered specialist employees that can operate on autopilot with their own objective function, taking on the core responsibilities of key departmental functions.
Take HR and talent acquisition as an example. The typical tasks of a sourcer can now be delegated to four main agents. AI search agent can autonomously scan vast talent pools on platforms like LinkedIn, identifying high-fit candidates using nuanced criteria that go far beyond simple keyword matching - analysing skills, experiences, and even inferred potential.Evaluators use AI to assess these candidates’ suitability while anonymising profiles and employing fairness checks to mitigate bias. Outreach agents handle engagement at scale, creating hyper-personalised messages that resonate with candidates, and scheduling agents seamlessly coordinate interviews by dynamically managing calendars across teams.
—Teo Borschberg, CEO at Ogment Inc
Advancements in data infrastructure…
In 2025, I am excited about advancements in data infrastructure enabling AI by unifying fragmented systems and streamlining ETL. Tools like Apache Iceberg, Delta Lake, and Trino reduce latency, while metadata-driven orchestration ensures consistent, high-quality data for AI frameworks like PyTorch and TensorFlow.
These innovations bridge legacy and cloud systems, enabling faster, more effective AI deployment for enterprise use cases like fraud detection and predictive analytics.
—Chris Zeoli, Partner at Wing Venture Capital
Building products at the fraction of a price…
In 2025, I am excited about the seismic shift that AI will bring to personal outcomes. From using AI coders to create apps and monetize them within seconds, to using AI language learning tutors or live translation to remove cultural barriers, to 10x-ing performance at work with the use of agents. The cost of creating something from scratch has gone down tremendously and we will see exciting new products pop up at a faster pace than ever.
—Mila Cramer, Principal Project A Ventures
Democratizing AI through open source technologies…
In 2025, I am excited about the broad democratization of AI through open source technologies. Companies can now build and deploy specialized models using mature, open source tools rather than relying on external proprietary solutions.
This shift is unlocking a new wave of practical AI adoption, as organizations of all sizes can finally create AI tailored to their specific needs without sacrificing control over their production data.
—Misha Herscu, CEO and Co-Founder Cake AI
AI becoming more intuitive and conversational…
In 2025, I am excited to see interactions with AI become more intuitive and conversational, moving away from clunky and transactional exchanges. As voice and video models become more capable I believe we will see a large shift in adoption toward voice being the dominant way we communicate with AI in our daily lives.
—Matthew Hennigan, Senior Associate at Plug and Play Ventures
LLMs to run on CPUs of regular laptops…
In 2025, I am excited about small, local LLMs most likely becoming small and efficient enough to run on CPUs of regular laptops, while being competitive with the hosted LLMs such as OpenAI's, at least for specific tasks.
—Egor Kraev, Principal AI Scientist at Wise Plc
AI shaping more personal and private experiences…
In 2025, I’m excited about AI becoming more integrated into people’s daily lives - feeling less like a service they need to invoke at times - and more an intuitive force that somehow always makes their work and life more productive and joyful.
I’m also excited about AI moving closer to the edge on people’s devices allowing us to shape more personal and private experiences.
I also remain optimistic about the breakthroughs in the physical sciences particularly medicine and material science that it has the potential to unlock.
—Satyajeet Salgar, Director of Product Management and User Experience, AI Innovation and Research at Google
Chain of thought reasoning models…
In 2025, I am excited about chain of thought reasoning models. LLMs are largely considered a black box which takes in inputs and produces an output. However, there are a lot of reliability issues with this type of model, as we have no easy way to actually understand how the model was able to arrive at its final output.
However, with chain of thought models we can see the model start to "reason" similar to how a human would, and partially understand how it arrived at its answer. This ability to deconstruct a complex problem into its constituent pieces will be crucial as we develop more comprehensive agentic models, and try to make models more reliable.
—Rahul Gudise, CEO at Gale
The takeover of horizontal agents…
In 2025, I am excited about the takeover of horizontal AI agents. These are true generalists who can learn and excel at a broad range of tasks. In contrast, what today are called "agents" often reduce to vertical-specific "pipelines". With increasingly better reasoning models and multimodal perception, 2025 will be the year horizontal agents flourish.
—Harvey Hu, Founder and CTO at Tessa
A new era for travel search…
In 2025, I am excited about LLMs and AI Agents to redefine the travel research, planning and booking process, making it a highly contextual, personalized and super easy agentic workflow.
Access to truly personalized insights and recommendations, based on trusted content, comprehensive 1st party data and deep usage of Generative AI, will be game changing for millions of global travelers, making adventures easier to plan and opening doors to incredible new experiences.
—Rahul Todkar, Head of Data and AI at Tripadvisor
AI voice agents sound remarkably human…
In 2025, I’m excited about the next wave of AI voice agents, especially their ability to sound remarkably human. What’s truly amazing is how capable large language models (LLMs) have become, now able to deliver hyper-accurate answers, even to highly specialized questions, with near-perfect precision.
That kind of precision and realism is a game-changer for customer support roles, where deep expertise on a specific system can make all the difference.
—Shaun Lane, Co-Founder and CEO at Riviera
AI safety…
In 2025, I am excited about AI Safety. As we head closer to AGI, AI safety becomes more important than ever, but at the same time, getting alignment is difficult.
—Jeffrey Ip, Co-founder and CEO at Confident AI
Enhancing sustainability…
In 2025, I’m most excited about AI-driven technologies that enhance sustainability by empowering smarter decision-making across industries. In agriculture, for example, AI is transforming how we monitor and optimize plant health, reduce waste, and improve resource efficiency, aligning with the urgent need for sustainable food systems.
Beyond agriculture, AI is driving innovation in renewable energy management, circular economy solutions, and climate modeling, helping us tackle global challenges with actionable insights. What excites me most is how these technologies are moving us closer to a future where innovation and sustainability are inseparable, creating tangible solutions for both people and the planet.
—Valeria Kogan, CEO at Fermata
READY TO BUILD? LEARN FROM THE BEST
We have put together some of the most forward thinking AI companies including their pitch decks, all accessible once you become part of our community—as a paid subscriber.
BECOME A PAID SUBSCRIBER TO ACCESS THE BEST IN CLASS AI PITCH DECKS NOW.
Companies include:
ElevenLabs — Pioneering research in Text to Speech, AI Voice Generator, and more.
Vercel — Provides the developer tools and cloud infrastructure to build, scale, and secure a faster, more personalized web.
Mistral AI — Frontier AI. In Your Hands.
Perplexity — A conversational search engine.
Orbital Witness — The all-in-one property diligence automation platform for lawyers and property professionals.
Gendo — AI for Architecture & Design.
Wondercraft — Amazingly easy audio content creation.
Kinnu — Learn anything. Remember everything.
Portkey — Control Panel for AI Apps
Elise AI — Industry Leading AI for Housing and Healthcare
Aindo — Trusted synthetic data.
Supernormal — Make every meeting count.
Creative Fabrica — Premium Crafting Fonts, Graphics & More.
Hour One — Human-Centric Storytelling.
Synthesia — Turn text to video, in minutes.
Qventus — AI teammates to automate hospital operations.
Reflexivity — Data that finally speaks your language.
StackGen — The Generative Infrastructure Platform.
RapidCanvas — Trusted AI solutions that deliver measurable business impact fast.
Endeavor — Grow, keep, & win more customers with AI Agents.
Pruna AI — Make your models faster.
Deckmatch — AI Workflows for Your Chaotic Top of Funnel.
Discover how you can become the next big AI startup now.
If you have enjoyed reading our annual AI report, please help sustain our work by clicking ❤️ and 🔄 at the top of this post.
Once again, a huge thank you to our partners!