Can AI do AI Research?

AI research can be carried out entirely within digital spaces, making it ripe for automation. Recent efforts have demonstrated that AI systems are capable of carrying out the whole process of research from ideation to publishing. Startup Sakana.ai has created an 'AI Scientist' that independently chooses research topics, conducts experiments, and publishes complete papers showing its results. While the quality of this work is still only comparable to an early-stage researcher, things will only improve from here.

Judging Social Situations

AI chatbots, including Claude and Microsoft Copilot, can outperform humans in evaluating social situations. In an established 'Situational Judgment Test', these AI systems consistently selected more effective responses than human participants.

SOURCE

Analyzing Scientific Literature

While language models are known to hallucinate information, this tendency can be reduced. PaperQA2, an LLM optimized to reliably provide factual information, was able to match or exceed human subject matter experts across a range of realistic literature review tasks. The summary articles it produced were found to be more accurate than those written by human authors.

SOURCE

Writing Emotive Poetry

A study has shown that non-expert readers can no longer tell AI-authored poems from those written by acclaimed human poets. The AI poems were also rated higher in rhythm and beauty.

SOURCE

Writing Post-surgical Operative Reports

Surgeons take painstaking notes of the actions they carry out during surgeries, collecting them into narrative form as an 'operative report'. A machine vision system was trained to watch surgery footage and produce such reports. It did so with higher accuracy (and much higher speed) than human authors.

SOURCE

Developing New Algorithms

AIs can find innovative solutions to difficult coding problems when given an appropriate framing. For example, a dedicated system called AlphaDev was trained to play a game about creating sorting algorithms. The algorithms it discovered were novel and outperformed existing human-authored benchmarks.

SOURCE

Who is Building AGI?

The following companies have explicitly stated they intend to develop AGI, either through public statements or in response to FLI’s 2024 AI Safety Index survey:

Anthropic

OpenAI

Google DeepMind

Meta

x.AI

Zhipu AI

Alibaba

DeepSeek

How can we avoid AGI?

There are policies we can implement to avoid some of the dangers of rapid power seeking through AI. They include:

Compute accounting
Standardized tracking and verification of AI computational power usage

Compute caps
Hard limits on computational power for AI systems, enforced through law and hardware

Enhanced liability
Strict legal responsibility for developers of highly autonomous, general, and capable AI

Tiered safety standards
Comprehensive safety requirements that scale with system capability and risk

TOMORROW’S AI

ALL SCENARIOS

Clear Sight

BACK TO SCENARIOS

Intended Use: Corporate/Finance

Technology Type: Problem-Solving/Cognitive

Runaway Type: Concentration of Power

Control Lever: Transparency and Accountability

Primary Setting: EU

The Trouble With Taxes

By the late 2030s, corporate tax avoidance reaches historic levels. Multinational firms leverage AI-driven tax strategies faster than regulators can react. The rise of gig work, decentralized finance, and influencer-driven businesses leaves governments struggling to track the taxable income of individuals. An estimated €350 billion in tax revenue is lost annually in the EU alone, exacerbating wealth inequality and straining public services.

ClearSight

Facing mounting public pressure, the European Commission launches ClearSight: a tool AI forensic auditor designed to expose hidden wealth, close loopholes, and restore fiscal fairness. It is designed to have high explainability and transparency. Corporate lobbies and financial elites fight back, warning of “algorithmic surveillance.” But voters demand action, and ClearSight goes live.

Preemptive Policymaking

ClearSight combines machine learning, anomaly detection, and blockchain analytics to scan billions of financial transactions daily. It flags unusual filing patterns, inconsistent declarations across jurisdictions, and mismatches with publicly available data. Human auditors remain central to enforcement, using ClearSight’s insights to trigger investigations and propose legislative fixes to newly identified loopholes.

Turning the Tide

In just three years, corporate tax compliance rises by 28%, recovering €120 billion for public infrastructure and social programs. Fraud scandals erupt as multiple highly-valued startups and family offices are exposed for hoarding untaxed billions. Meanwhile, the ultra-wealthy deploy private tax AIs to stay one step ahead, weaponizing preemptive asset restructuring. Corporate tax havens like the Cayman Islands and Singapore also resist full AI integration, allowing some firms to continue shifting assets offshore.

Evasive Maneuvers

As AI audits expand, new black-market financial networks emerge. Underground economies turn to cash, crypto, and barter systems to evade automated tracking. Some gig workers deliberately underreport their income, believing that AI tax enforcement is disproportionately targeting digital freelancers while corporate giants game the system. Facing backlash, regulators adjust ClearSight’s scope, focusing AI enforcement on high-income tax evaders and making enforcement tools accessible to small businesses and freelancers.

A Fairer Future

Despite the opposition, ClearSight fundamentally reshapes global taxation. Its transparent architecture and audit trail systems earn international trust, helping to standardize enforcement across borders. Over time, cooperation grows into a universal AI tax framework that ensures corporations, digital laborers, and financial elites all contribute fairly. For the first time, tax policy is not shaped by those who benefit from its flaws, but enforced by a system built to protect the public good.

BACK TO SCENARIOS