Big Tech can’t own AI’s future if we decentralize it first

By admin On Nov 14, 2024

While AI’s potential to benefit humanity is vast, its trajectory under corporate monopolies threatens to concentrate power, excluding most stakeholders from shaping its future. The best way forward? Decentralize AI through community-driven, blockchain-based models and build a transparent, inclusive ecosystem that keeps AI safe, ethical and accessible to everyone.

When OpenAI launched in 2015, it aimed to “benefit all of humanity” with an open source, nonprofit approach. Yet, less than a decade (and much leadership turmoil) later, its pivot to a for-profit model — driven by a pressing $6.6 billion debt to investors — has made it clear that corporate interests, not collective benefit, are increasingly shaping AI’s development. The move illustrates a wider trend: Today’s leading AI organizations are becoming tightly controlled, profit-driven and increasingly opaque.

While these changes might be justifiable from a business standpoint, AI’s future shouldn’t rest solely with a few powerful companies. The shift toward private investor-driven AI has the potential for dystopian consequences, excluding grassroots users and developers from contributing meaningfully, just as many were sidelined in the Web2 era.

Read more from our opinion section: A small group of tech giants are holding AI’s future hostage

To counterbalance this trend, decentralized and community-led AI initiatives are crucial. These would offer an alternative where AI systems remain accessible, safe and unbiased, reducing the impact of profit-driven “black box” models.

Today, AI development is dominated by a few major players — OpenAI-Microsoft, Google, Amazon, Meta, Nvidia — all racing toward monopolization. By controlling critical parts of the AI stack — models, training data and applications — companies can maximize profits, keeping revenues and innovation tightly siloed.

Moreover, cloud infrastructure controlled by Amazon and Google provides most of the compute and storage resources needed to train and deploy AI systems. This centralization comes at a steep cost: According to Akash Network, grassroots developers will spend over $679 billion on cloud computing in 2024 — already unaffordable for most small-medium players, with costs likely to rise as demand grows.

Within Big Tech’s walled gardens, users are given limited, conditional access to AI tools and APIs — often at exorbitant prices. Because AI’s development has become a winner-takes-all game, corporate interests will continue to compromise long-term ethical considerations in favor of immediate revenue. OpenAI’s recent dissolution of its Superalignment team, as Jan Leike pointed out, suggests that safety and accountability are taking a backseat to high-profile product launches.

Decentralization is the best antidote to BigAI monopolies. Ongoing blockchain AI innovations let user and developer communities become active, incentivized contributors in AI’s evolution. Projects like KIP Protocol have pioneered decentralized ownership of the AI stack, allowing individuals to contribute data for model training without needing to build the models themselves. This lowers entry barriers, while crypto-based payment rails enable non-intermediated profit-sharing among stakeholders.

Read more from our opinion section: The internet is broken. What if we can fix it?: On Chris Dixon’s ‘Read Write Own’

Similarly, Akash Network’s decentralized marketplace provides storage and compute resources at roughly 80% lower costs than centralized providers. Projects such as Flock.io are already leveraging Akash’s infrastructure for decentralized AI training.

Beyond these examples, decentralized physical infrastructure (DePIN) projects are one of DeFi’s most active areas, according to Nansen. With initiatives like NodeFi, end-users can monetize excess compute or storage on personal devices. By offering governance rights to node owners, projects in this space can take community-owned AI to the next level.

Funding trends reflect this momentum. Decentralized AI projects are drawing attention from investors, with AI startups raising 30% of total funding in Q3 2024 — an impressive $52.2 billion, according to Stocklytics. Blockchain AI ventures like Wire Network and OpenGradient have emerged alongside traditional players, and the gap between decentralized and centralized AI presents immense opportunities for growth.

This influx of funding is more than just a financial trend; it signals a shift in industry priorities towards decentralized alternatives. Decentralized AI offers a feasible, attractive alternative to BigAI, promising an inclusive, equitable and accessible future. BigAI is moving at breakneck speed to establish dominance, and the time to support decentralized alternatives is now. There is no looking back.

Source

0 20

Share FacebookTwitterGoogle+ReddItWhatsAppPinterestEmail