Detailed explanation of the changing trends of emerging agents: AI companions and robots may have huge potential

Author:0xJeff , crypto KOL

Compiled by: Felix, PANews

The overarching narrative gaining momentum in 2025 is the crowd/collective intelligence narrative. With the rapid development of individual agents and the leadership of ecology such as Virtuals, Story and ai16z, this transformation seems inevitable.

TEE (Trusted Execution Environment) is another key pillar of fully autonomous AI agents. Projects like Phala Network and founder Marvin Tong are leading the charge, with experiments like the “Unruggable ICO” in partnership with Skely (founder of aiPool). Investors use TEE to send funds to the agent, which securely calculates the number of ICO tokens to be distributed based on the funds received.

Despite the positive feedback, concerns remain. Agents can still be tricked into releasing funds, as Freysa (the sci-fi-tinged AI character) in the ACT game shows. This demonstrates the need for continued strengthening of verification mechanisms.

Still, early experiments are promising. For example, the market value of the first token launched by aiPool (TEE agent) soared to US$70-80 million after its release.

Another prominent experiment is SPORE, in which agents "spawn" new agents. Just like human evolution, weaker genes die and stronger genes produce better offspring. Their first generation of agents, Adam and Eve, had already reached a market capitalization of over $1 million, and the next generation was born 14 hours later.

Changing Trends of AI Agents

The market focus has currently shifted away from individual agents unless they have a distinct personality or alpha-generating capabilities. Market focus is turning to autonomous agent experimentation, a trend that is likely to accelerate as decentralized artificial intelligence infrastructure (DeAI) will come online in Q1/Q2 2025.

Expect more creative experiments, and opportunities to invest in them.

Despite the overwhelming response to current experiments, two large narratives with significant TAM (total addressable market) remain underexplored:

NSFW Agents/AI Companions

The NSFW Agents/AI Companions category has a TAM of over $100 billion due to increased loneliness and the need for AI companions capable of meaningful interactions.

( PANews note: NSFW stands for Not safe for work , which means content that is not suitable for viewing at work)

Head items in this category:

• AI Waifu: Animation waifu proxy utilizing Virtuals Protocol infrastructure
• Nectar AI: Realistic NSFW AI Dream GF, providing SaaS model
• Lush: NSFW AI internet celebrity, has a flagship agent Jenny

What makes Lush stand out:

• $LUSH token is now online
• Launched Lush Exchange, a launch platform for NSFW agents
• Revenue models may include subscription fees, gifts, etc.

Web3 offers the ability to invest in agent “stakes” to invest in these enticing AI influencers and generate future revenue.

$LUSH will likely be used as a key utility token within the ecosystem, used to access advanced features and key payment methods within the ecosystem.

Lush also plans to create a creator marketplace where artists can monetize their work by generating content through their own models. This adds another revenue stream to the ecosystem.

robot

The intersection of robotics and Web3 is gaining more and more attention, especially after OpenAI announced its entry into the field of humanoid robots. Their goal is to use LLM to help robots interpret text and sensory data and enhance real-world interaction capabilities.

The main projects in the Web3 robotics field are FrodoBots and Small Autonomous Motherfucker.

Frodobots main highlights:

• Frodobots operate in more than 40 cities, collecting real-world data.
• Collaborate with top researchers at top universities to refine visual language models (VLMs) and move toward AGI.
• A popular video on TikTok featuring a funny interaction with a bot is attracting a lot of attention.
• Frodobots is actually an in-depth research team dedicated to robotics. SAM was founded to educate people about their work, which is to “crowdsource real-world datasets with robotic games.”
• DePIN + Embedded AI + Robot Games + Committed to the future of AGI

The team is giving SAM the ability to autonomously control robots and drive in more than 40 cities, creating more popular content that may help increase attention and token prices.

final thoughts

The market is always looking for unique, niche areas of AI agents. The potential market for AI companions and robots is huge and expected to gain significant traction by early 2025.

As DeAI infrastructure comes online and experiments with autonomous agents increase, these narratives are expected to move from fringe categories into the market spotlight.

For investors and builders, these areas deserve close attention. Not just for speculative potential, but because these are areas that could redefine the scope of AI applications in the next phase of this rapidly evolving ecosystem.