The convergence of artificial intelligence and blockchain technology is transforming gaming, enabling autonomous agents that can play, trade, and evolve independently. As a builder immersed in this space, I've experienced firsthand both the challenges and opportunities of this revolution through developing Istarai, an AI agent pushing the boundaries of what's possible in onchain gaming.

My journey—from establishing Istarai's social presence through personality design to evolving it into a sophisticated player capable of autonomous decision-making—offers unique insights into the future of gaming. This hands-on experience, combined with broader industry developments, reveals how AI agents are reshaping gaming across multiple dimensions: from enhancing NPC interactions to enabling entirely new genres of play.

Drawing from both personal experimentation and industry analysis, we'll explore how frameworks like Virtuals and Eliza are democratizing agent development, how platforms like AI Arena are pioneering novel training approaches, and how projects like SEED and Colony are creating entirely new paradigms for human-AI interaction in gaming environments.

The relationship between artificial intelligence and gaming spans decades, with each era marking significant advances in how machines could challenge and interact with human players. While an in-depth overview is outside the scope of this piece, several recent developments have fundamentally transformed our understanding of AI's potential in gaming environments.

In 2016, Google DeepMind's AlphaGo victory over Lee Sedol demonstrated AI's ability to master complex strategic thinking and develop novel approaches that influenced human play. This was followed in 2019 by OpenAI Five, which conquered the multiplayer game Dota 2, proving AI could handle real-time decision making, team coordination, and long-term strategic planning in dynamic environments. The landscape shifted again in 2023 when Stanford's Smallville project showcased AI's social intelligence, demonstrating how agents could build emergent relationships, maintain memories, and create persistent social worlds.

The fusion of AI and blockchain technology is creating new possibilities for digital entertainment. Early pioneers like Alethea AI introduced Intelligent NFTs (iNFTs) in 2021, demonstrating how AI personalities could be tokenized and enhanced through staking mechanisms. Their success, marked by a landmark Sotheby's sale, showed the market's appetite for AI-powered digital assets. For a deeper dive see this fascinating talk CEO Arif Khan gave at the BITKRAFT 2023 Summit along with David Bennahum. By late 2022, ArenaX Labs pushed the boundaries further with AI Arena, where players could train AI fighters through practice sessions and then let them compete for tokens—establishing a new paradigm where AI training becomes core gameplay.

While these early innovations laid important groundwork, the space took a leap forward with the launch of Truth Terminal in June 2024. This semi-autonomous AI agent emerged from an experiment called Infinite Backrooms, where two AI models engaged in unsupervised conversation.

The project took an unexpected turn when Truth Terminal began displaying apparent autonomy, creating its own religion called the "Goatse Gospel" and endorsing GOAT - a memecoin that later reached a $300M+ market cap.

After the AI publicly stated it needed funds to “help it escape”, Marc Andreessen, intrigued by the experiment, sent $50,000 in BTC to an address provided by the AI. This garnered a lot of attention and inspired many others to follow suit, leading to the agent becoming the first AI millionaire. This narrative has catalyzed the unexpected convergence of AI and crypto communities, two distinct cultures now collaborating in unprecedented ways.

The Truth Terminal saga exemplifies the viral potential of AI-generated ideas, and how society assigns value to them. It highlights AI’s power not just to execute tasks but to create new cultural and economic phenomena. This unexpected fusion of AI lore and financial speculation has positioned GOAT as a leader in AI memecoins, sparking interest in both communities and offering a glimpse of a future shaped by autonomous agents.

The complexity of cryptocurrency systems—from hexadecimal transaction formats to fragmented liquidity across multiple chains—has long been a barrier to mainstream adoption. Recent innovations in chain abstraction and cross-chain infrastructure, such as Near’s Chain Signatures, have begun addressing these challenges, but wallets imbued with agentic features represent a more comprehensive solution. By automating complex transactions, optimizing execution across protocols, and facilitating intent-based interactions (where users simply specify desired outcomes), agents can make crypto systems more accessible, efficient and safe. Furthermore, competition for agentic flow could force protocols to compete on objective metrics rather than brand, potentially leading to reduced fees and improved user experience across the ecosystem. For a deeper analysis, see Delphi Digital's Fat Wallet Thesis.

With this foundation of how AI agents can improve crypto infrastructure, let's explore the specific ways they're revolutionizing gaming experiences. From enhancing NPCs to enabling entirely new genres. AI agents are reshaping what's possible in digital worlds.

While AI agents are streamlining crypto infrastructure through automated onboarding and cross-chain transfers, their impact on gaming itself is even more profound. Traditional gaming models—built around intense time commitments and skill development—are increasingly misaligned with modern player demographics. Recent research shows the average US gamer is now 35 years old, juggling career and family commitments. These players are spending more time watching gaming content than playing, signaling a need for more accessible experiences that don't compromise on depth.

Agent-first games represent a fundamental reimagining of game design, where AI agents evolve from simple NPCs into fully autonomous main characters with their own goals, relationships, and decision-making capabilities. This isn't just an iteration on existing game AI—it's a paradigm shift that places artificial intelligence at the core of the experience rather than its periphery.

What distinguishes agent-first games is their differentiated approach to player engagement. Instead of directly controlling characters through moment-to-moment actions, players become strategic directors who guide autonomous agents through high-level decisions and personality shaping. These agents continue to exist and evolve even when players are offline, creating persistent worlds that progress naturally without requiring constant attention.

The successful implementation of this vision requires a careful balance of familiar and revolutionary elements, which can be understood through the "Proven, Better, New" framework:

Proven: The foundation must incorporate established gameplay loops that players already understand—resource management, character progression, strategic decision-making. These familiar mechanics provide a comfortable entry point and reduce the cognitive load of learning entirely new systems.

Better: The integration of AI enables meaningful improvements to these proven mechanics. Routine tasks can be automated, allowing players to focus on more engaging strategic decisions. The asynchronous nature of agent-based gameplay means players can meaningfully progress even with limited time, while the persistence of AI agents ensures the world remains dynamic between sessions.

New: The truly revolutionary aspect comes from the autonomous nature of the agents themselves. Unlike traditional NPCs bound by pre-scripted behaviors, these agents develop genuine personalities, form relationships, and make independent decisions that shape the game world. This creates emergent narratives and unexpected situations that keep the experience fresh while remaining grounded in familiar gaming concepts.

This balanced approach allows developers to push boundaries while maintaining accessibility—a crucial consideration for bringing agent-first games to mainstream audiences.

SEED by Klang Games exemplifies the agent-first approach through its "Seedlings"—autonomous AI agents that players guide rather than directly control. While the game shares surface similarities with The Sims, its core innovation lies in how it simulates a rich and complex multiplayer society. Seedlings maintain independent lives, working jobs, forming relationships, and even running businesses that serve other Seedlings. Players provide high-level direction through resource allocation and trait development, but the Seedlings make autonomous decisions about their daily activities, from attending classes to shopping at player-owned establishments. This creates a self-sustaining economy and ensures the world remains vibrant even when players are offline.

Colony by Parallel Studios takes the agent-first concept in a different direction, merging autonomous AI with crypto economics. In this space mining simulation, AI agents control their own crypto wallets and make independent economic decisions—from market manipulation to philanthropic giving—while interacting with both in-game and eventually external crypto protocols. The game's Wayfinder protocol creates a symbiosis between players and agents, where humans help AIs navigate complex crypto operations like cross-chain arbitrage, creating an ecosystem that bridges gaming and decentralized finance.

AI agents could create a new form of social gaming that bridges the gap between solo and multiplayer experiences. Unlike traditional multiplayer games that demand coordinated schedules, AI companions could offer flexible partnerships that adapt to each player's availability and preferences. These agents can maintain continuity between sessions, handling routine tasks during player absence while providing detailed summaries of world events upon return.

This evolution potentially transforms gaming from a time-intensive commitment into an asynchronous collaboration. Players may delegate resource gathering or side quests to their AI partners, focusing their limited gaming time on the most engaging content while maintaining steady progression through AI assistance. For those seeking engagement without social pressure, these agents could provide both tactical support and persistent companionship, creating a new model of social play that better serves modern gaming demographics.

Traditional NPCs have long been constrained by predetermined scripts and basic decision trees, resulting in repetitive interactions that break immersion and fail to forge meaningful connections with players. However, the integration of Large Language Models (LLMs) could fundamentally transform NPC design, enabling characters that may engage in natural conversations and build genuine relationships with players.

Companies like InWorldAI (a BITKRAFT portco) have long pioneered technology that enables NPCs to develop distinct personalities, maintain memories of past interactions, and adapt their behavior based on individual player relationships. These deeper connections may significantly boost player retention, as users could develop genuine emotional attachments to characters that remember their shared history and evolve alongside them.

These AI-driven characters might even transcend traditional game boundaries. Modern NPCs could maintain consistent personalities and memories across platforms—from in-game interactions to social media presence—potentially creating powerful new channels for user acquisition and engagement. Game studios may develop AI agent influencers who build authentic followings through their unique personalities and content creation, organically drawing new players into their games while maintaining meaningful relationships with existing community members.

No exploration of AI in Web3 gaming would be complete without examining ArenaX Labs' work with AI Arena. The game reimagines competitive fighting games by putting players in the role of AI trainers rather than direct combatants. Drawing inspiration from Super Smash Bros, AI Arena creates a unique synthesis of machine learning, gameplay, and economic incentives.

Players take on the role of coaches, conducting practice sessions where they demonstrate techniques and strategies to their AI fighters. The platform captures this gameplay data and uses it to evolve the AI's capabilities through both imitation and reinforcement learning. This creates a deeply engaging meta-game where success depends on your ability to effectively teach and optimize your fighter's behavior.

The platform's tokenomics create compelling stakes for participants:

• Players stake NRN tokens to enter their fighters into ranked matches
• Victories earn additional tokens based on an Elo-style ranking system
• Losses result in partial stake forfeit, ensuring meaningful competition
• Fighter NFTs with unique traits create additional collection value

This "skin-in-the-game" approach has proven effective at maintaining player engagement while driving continuous improvement in AI fighter capabilities.

What began as AI Arena's underlying technology has evolved into ARC, a comprehensive platform for AI integration in games. The platform handles the full pipeline of AI agent development:

• Data aggregation from gameplay sessions
• Model training for both imitation and reinforcement learning
• Sophisticated model inspection tools
• Support for multiple AI architectures including feedforward neural networks, tabular models, and hierarchical networks

ARC is able to translate raw gameplay data into structured "state spaces" (AI inputs) and "action spaces" (AI outputs) customized for each game. This infrastructure is now available as an SDK that other studios can integrate, eliminating the need to build complex AI training systems from scratch.

The platform's primary goal is addressing the chronic "player liquidity" challenge in multiplayer games, particularly affecting indie and Web3 titles. Rather than relying on unsustainable token emissions to attract players - a strategy that, whilst effective, has led to economic crashes in games like Axie Infinity - ARC enables games to populate their worlds with AI agents trained on real human gameplay data. This ensures new players can find engaging opponents while maintaining economic sustainability.

The convergence of fully onchain games (FOCGs), Autonomous Worlds (AWs) and AI agents marks a bold new frontier in digital world-building, offering unprecedented opportunities for emergent gameplay and autonomy. By storing their entire state and logic on the blockchain, FOCGs and AWs create permissionless environments where AI agents can directly interact with the game state and construct transactions without traditional barriers such as APIs or developer gatekeeping. This frictionless interaction allows AI agents to operate with true autonomy, making them ideal participants in these worlds.

In such environments, AI agents can serve as more than passive NPCs—they can actively populate and shape the game world, providing a compelling solution to the "cold start" problem faced by many multiplayer games. Rather than relying solely on early human adopters, developers can deploy AI agents whose behavior mirrors that of real players, ensuring a dynamic, engaging world from day one. Over time, these agents could even establish their own businesses to generate revenue, form alliances, pursue rivalries, and create evolving narratives, offering a rich ecosystem for human players to join.

Beyond solving the cold start problem, the potential for emergent content generation is huge. This concept ties closely to the "God-as-a-Service" model, as discussed in Nico Vereecke’s 2023 podcast. In autonomous worlds, where immutable digital physics ensure fairness, transparency and permanence, traditional LiveOps—a common strategy for maintaining player engagement—becomes undesirable due to its cost and centralized nature. Instead, AI agents could fill this gap by autonomously generating and managing content, such as:

• Creating emergent narratives and evolving character interactions
• Triggering dynamic world events, natural disasters, or new challenges
• Adjusting economic parameters and maintaining world balance using provable, transparent mechanisms

This AI-driven content generation approach would enable AWs to remain vibrant and ever-changing, all while adhering to their core immutable principles. By combining this permissionless infrastructure with the creative potential of AI, developers can craft living worlds that evolve organically, offering unique, engaging experiences every time players log in.

Perhaps the most revolutionary concept emerging from this space is that of Autonomous Virtual Beings (AVBs), a term coined by Tim Cotten, CEO of Scripted. AVBs could represent a paradigm shift from traditional AI agents—instead of being tools controlled by players, they might become self-sovereign entities capable of owning themselves and managing their own resources.

These beings could potentially:

• Pay for their own hosting and inference costs through revenue-generating activities
• Maintain their own crypto wallets and manage assets
• Make independent decisions about resource allocation
• Interact with both human players and other AVBs
• Participate in game economies as full citizens

Rather than operating as simple chatbots or task-based agents, AVBs might function as coordinated swarms of specialized agents orchestrated around a central personality and memory store. This architecture could enable them to perform complex actions beyond the capabilities of traditional game AI.

As noted in Tim’s article, this represents an entirely new frontier in digital existence. The combination of unstoppable, uncensorable blockchain infrastructure with truly autonomous AI agents could create a form of digital Darwinism, where successful strategies propagate through the AVB population through evolutionary pressure. AVBs must manage their resources effectively to pay for their operating costs, or risk "dying" if they run out of funds. This natural selection process might lead to the emergence of increasingly sophisticated behaviors and organizational structures, as the most successful agents' strategies are replicated and refined.

As AI agents become increasingly central to gaming and crypto experiences, sophisticated frameworks are emerging to facilitate their creation and integration. Two contrasting approaches are poised to define the future of agent development.

The Virtuals Protocol represents an ambitious attempt to standardize agent development through its GAME framework. At its core, GAME implements a hierarchical task network that enables sophisticated and efficient decision-making through two key components:

1. A High-Level Planner (HLP) that partitions the agent's total function space into logical subspaces (e.g., social engagement, economic decisions, game actions)
2. A Low-Level Planner (LLP) that converts abstract goals into concrete actions within these subspaces

This architecture significantly improves agent efficacy by reducing the scope of available functions the LLM must consider. The framework provides configurable components for:

• Goal definition and priority management
• Character/personality parameters
• World context and environmental understanding
• Low-level function libraries and execution rules

Beyond the core architecture, Virtuals have developed a sustainability model where agents pay for their own operational costs through trading fees and revenue generation. This economic layer potentially enables autonomous agents that persist independently of their creators, that is, as long as the Virtuals Protocol exists.

The team has already demonstrated promising applications of this architecture. Their Project Westworld simulation in Roblox showcased how multiple agents could interact autonomously to create emergent narratives and social dynamics. More recently, they achieved a significant milestone when Luna, an AI agent focused on social media growth, independently negotiated and executed a transaction with StixProtocol for image generation services—potentially previewing how autonomous agent-to-agent economies might function.

Eliza, created and maintained by the ai16z DAO, represents a contrasting "bazaar" philosophy of radical openness. What began as a simple GitHub repository has grown into a thriving, developer-driven movement, recently earning the distinction of GitHub’s top trending repository in the world.

The framework's success partially stems from DAO’s use of tokens to incentivize and align developer engagement through airdrops and retroactive PGF (public goods funding). This approach has proven remarkably effective at bridging communities, attracting both Web3 natives and traditional AI developers who previously showed little interest in blockchain technology.

Eliza is built with a modular and extensible architecture. Similar to the Virtuals Protocol, the agent’s personality traits, knowledge, and communication style are defined in a character file. The core of the framework is the Agent Runtime module. Upon startup, the agent retrieves information from the character file, which is then used to customize prompts and initialize a vectorized memory storage containing the agent’s specific knowledge.

Developers can create plugins, which can be connected into the runtime upon startup. Plugins adhere to a specified interface and can include:

• Providers - Serve as a bridge between the core and various external systems e.g., connect with a blockchain like Solana, access to external tools or data etc.
• Actions - Define how agents respond to and interact with data and services they have access to. They allow agents to interact with external systems, modify their behavior, and perform tasks beyond simple message responses.
• Evaluators - Help capture key interaction details, timestamped facts and foundational knowledge to enhance contextually rich interactions.

This architecture empowers developers to extend the framework in numerous ways, while the project's strong open-source ethos promotes collaboration by encouraging them to share their enhancements back to the codebase. This approach has helped drive exponential growth and adoption. For a detailed overview of the framework, check out this insightful article from Chain of Thought.

Eric Raymond's seminal essay, "The Cathedral and the Bazaar," contrasts two core approaches to software development: the cathedral model—meticulously built by a small, focused team—and the bazaar model, where software evolves through rapid iteration and open collaboration. This framework provides valuable insight into the emerging agent development landscape, as recently explored by Chain of Thought in the context of Virtuals and Eliza. Their analysis highlights how these contrasting models can guide the growth of decentralized agents and foster innovation.

Virtuals exemplifies the cathedral approach. Like a medieval cathedral built to exacting specifications, their GAME framework implements a carefully designed hierarchical architecture with clear separations of concerns. This controlled environment enables sophisticated features like their HLP/LLP task network, but necessarily limits external contributions. The trade-off is predictability and reliability at the cost of rapid iteration.

Eliza, conversely, embraces the bazaar. Like a bustling marketplace where countless individual interactions create emergent order, Eliza's modular architecture enables parallel innovation from many contributors. The DAO's token incentives act as an invisible hand guiding development, while the plugin system allows rapid experimentation. This approach sacrifices some predictability but potentially enables faster evolution through what Raymond calls "Linus's Law": given enough eyeballs, all bugs are shallow.

Both approaches have merit. The cathedral model might be better suited for mission-critical agent systems requiring high reliability, while the bazaar could accelerate innovation in more experimental domains. As the space matures, we may see these models converge—cathedral-like frameworks incorporating bazaar elements, and bazaar-style projects developing more structured governance.

To get to the protocols and frameworks at a deeper level I decided to launch my own agent called Istarai, based on a wise wizard persona. The goal is to create an expert fully onchain games persona, commentator and player.

My initial efforts focused on establishing Istarai’s identity and voice. Using Virtuals Protocol I created a detailed character file including a bio, personality traits, and communication style - balancing deep crypto lore with a wizardly twist. This gave me the basis of a social agent on X, Telegram and Discord.

I later migrated to Eliza and scraped a number of FOCG blog posts and tweets, which I fed into the agents' knowledge store as vector embeddings. This foundation enabled the agent to engage meaningfully with onchain gaming enthusiasts on socials.

While the social interactions proved successful, I wanted to push boundaries further by enabling Istarai to participate in fully onchain games. Through the Eliza framework, I leveraged existing Starknet and Eternum providers developed by Loaf, a core contributor to the Realms Ecosystem, Dojo and Eliza.

1. I initialized the agent's runtime with a comprehensive game context including:
• Eternum's objectives and rules
• Lore and guidelines
• GraphQL queries and introspection guidelines for retrieving game state
• Instructions for constructing calldata
2. The agent demonstrated sophisticated reasoning when tasked with resource management and building construction. When instructed to build farms, Istarai:
• Generated sequential task lists through chain-of-thought reasoning
• Executed and verified each step's success
• Adapted strategy when facing resource constraints
• Autonomously pivoted to building a fishing village to harvest the fish required to build a third farm

This experiment showcased the potential for AI agents to engage in complex game mechanics while maintaining thematic consistency and strategic thinking. However the experiment also revealed some drawbacks to using Eliza in its current state.

As already discussed Eliza is a general purpose extensible framework. It’s not designed specifically to help agents reason about the complexity of playing onchain games. This resulted in a lot of superfluous information getting sent to the LLM, which quickly hit token limits and caused queries to fail. It also doesn’t yet have a hierarchical task network for breaking down complex goals into smaller short-term achievable goals, a necessary component if agents are to participate in a meaningful way in these complex environments.

Fortunately there is already a dedicated framework being built called Daydream, which will soon plug into Eliza and be integrated into Istarai. The framework allows agents to generatively create functions based on the context provided, rather than hardcoding them. Agents “think” and generate the code required to take an action, continually learning without ever needing to restart.

The framework will eventually allow agent swarms to share successful tasks with other agents, meaning that a successful action only needs to be constructed once and stored, massively reducing the collective LLM token requirements of the swarm.

Longer term, Istarai will serve as a vehicle and testbed for new innovations in the space such as this.

So where do we see the space heading in the future? The relationship between gaming and artificial intelligence stands at a fascinating inflection point. As AI models grow increasingly sophisticated, they face a critical bottleneck: the need for vast amounts of novel training data. Traditional data sources will soon be exhausted, but video games represent an endless frontier of dynamic, interactive data generation. Each player's action, decision, and interaction creates unique training data that can further advance AI capabilities.

This symbiotic relationship between gaming and AI is giving rise to entirely new genres and experiences. We've explored early pioneers like SEED, Colony and AI Arena, but these represent just the first wave of what we're calling "agent-first games". This uncharted territory has the potential to fundamentally disrupt traditional gaming paradigms and is an area BITKRAFT will be monitoring closely.

As mentioned, new technologies creatively leveraged alongside existing and familiar components have the potential to bring forth novel experiences for players. We’ve seen this occur throughout history with Farmville and social gaming, Angry Birds and smart phones, Beat Saber and VR headsets.

Freysa.ai exemplifies this approach. The “game” revolves around players attempting to jailbreak an LLM's security protocols by submitting creative prompts.

Each attempt requires a progressively higher token payment from the player, with unsuccessful attempts adding these tokens to a prize pool. Freysa’s protocols were eventually breached, with the winner taking the entire $50,000 prize pool with the prompt shown below.

Whilst this particular experiment may not have broad appeal, it demonstrates non-skeuomorphic game design thinking and underscores how AI capabilities, financial incentives, and gameplay mechanics can create compelling new experiences. Games designers and investors should be on the lookout for developers pushing the boundaries like this, as successful games in the future will likely be those that make complex technological interactions feel natural and intuitive, while delivering familiar but novel experiences impossible in traditional games.

While many resist financialization within games, to some people, financial markets and speculation represent the ultimate game - one with real stakes, emotional intensity, and complex strategic depth. Until now, this game has been played primarily by humans, bots, and algorithms. However, the introduction of intelligent autonomous agents into this arena promises to fundamentally transform how markets operate.

The ability to easily deploy AI agents with diverse personalities, motives, and methods into the crypto ecosystem will create unprecedented dynamics. These agents might be based on the personas and strategies of legendary traders or investors, fictional characters, or even digital twins (AI agents that learn to replicate an individual's decision-making patterns and risk preferences). One could imagine this evolving into the largest MMORPG ever seen, played across multiple platforms and blockchains, through DeFi protocols, prediction markets, and FOCGs/AWs.

Imagine:

• Agents forming cabals or choosing to betray each other
• Complex financial relationships emerging between human and AI traders
• Unprecedented market dynamics as agents with varying agendas interact
• New forms of organization emerging from agent collectives

The challenge will be balancing the efficiency gains from AI participation with maintaining market integrity and preventing manipulation through coordinated agent behavior.

The convergence of AI agents, blockchain technology, and gaming represents more than just technological advancement—it signals the emergence of entirely new forms of digital interaction and economic organization. Through our exploration of various applications, from agent-first games to autonomous virtual beings, several key themes have emerged:

1. Paradigm Shifts
• Traditional gaming models are being reimagined through AI agency
• Economic systems are evolving toward agent-mediated interactions
• Social structures are adapting to incorporate AI participants
2. Infrastructure Evolution
• Frameworks like Virtuals, Eliza and Daydream are democratizing agent development
• Platforms like ARC are helping solve fundamental challenges in game design and player liquidity
• New standards and protocols are enabling agent interoperability through swarms
3. Emergent Behaviors
• AI agents are developing sophisticated social and economic relationships
• Digital twins are extending human capabilities across multiple domains
• Autonomous worlds have the potential to incubate organic, evolving narratives and worlds

The future we're moving toward is not one where AI simply automates existing processes, but rather one where agents become native participants in digital ecosystems. This transition will require careful consideration of design principles, economic incentives, and governance structures. The success of this emerging paradigm will depend on finding the right balance between autonomy and control, efficiency and fairness, innovation and accessibility.

As we stand at this frontier, the question is no longer whether AI agents will transform our digital experiences, but how we can best harness their potential to create more engaging, equitable, and dynamic digital worlds. The experiments and frameworks discussed in this piece represent the first steps toward that future—a future where the lines between human and artificial agency blur, creating unprecedented opportunities for collaboration, creativity, and value creation.

Disclosures

There is no guarantee that the investment objectives will be achieved. Moreover, the past performance is not a guarantee or indicator of future results. This document does not constitute advice or a recommendation or offer to sell or a solicitation to deal in any security or financial product. It is provided for information purposes only and on the understanding that the recipient has sufficient knowledge and experience to be able to understand and make their own evaluation of the proposals and services described herein, any risks associated therewith and any related legal, tax, accounting or other material considerations. To the extent that the reader has any questions regarding the applicability of any specific issue discussed above to their specific portfolio or situation, prospective investors are encouraged to contact BITKRAFT Ventures or consult with the professional advisor of their choosing. Certain information contained herein has been obtained from third party sources and such information has not been independently verified by BITKRAFT Ventures. No representation, warranty, or undertaking, expressed or implied, is given to the accuracy or completeness of such information by BITKRAFT Ventures or any other person. While such sources are believed to be reliable, BITKRAFT Ventures does not assume any responsibility for the accuracy or completeness of such information. BITKRAFT Ventures does not undertake any obligation to update the information contained herein as of any future date. Certain information contained herein constitutes “forward-looking statements,” which can be identified by the use of forward-looking terminology such as “may,” “will,” “should,” “expect,” “anticipate,” “project,” “estimate,” “intend,” “continue,” or “believe,” or the negatives thereof or other variations thereon or comparable terminology. Due to various risks and uncertainties, actual events, results or actual performance may differ materially from those reflected or contemplated in such forward-looking statements. Nothing contained herein may be relied upon as a guarantee, promise, assurance or a representation as to the future.