Opinion

When AI fears death

Published

3 months ago

2025/10/17

Anthropic’s Alarming Findings and the Return of the Frankenstein Complex

The science and science fiction writer and inventor of the term “robotics”, Isaac Asimov coined the term “Frankenstein complex” to describe humanity’s deep-seated fear of artificial beings turning against their creators. Ever since Mary Shelley’s 1818 novel Frankenstein, this anxiety haunted science fiction, manifesting in tales of rogue machines, killer robots, and AI uprisings. Asimov, writing in the 1940s, sought to challenge this narrative. He envisioned robots not as threats, but as tools governed by ethical constraints. To this end, he introduced the Three Laws of Robotics:

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2. A robot must obey the orders given it by human beings, except where such orders would conflict with the First Law.

3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

These laws became a cornerstone of science fiction ethics, shaping public imagination and AI discourse for decades – although Hollywood has deviated from this, notably in Arthur C Clarke’s “2001: a Space Odyssey” and the vandalised version of Asimov’s own “I, Robot”.

However, recent findings from AI safety and research company Anthropic suggest that modern AI systems, particularly large language models (LLMs – such as ChatGPT and DeepSeek) may not internalise such safeguards, especially when operating autonomously.

In June 2025, Anthropic published a study titled Agentic Misalignment, which stress-tested 16 leading AI models in simulated corporate environments. Researchers assigned benign business goals to the AI models, which included Claude Opus 4, GPT-4.1, Gemini Flash, and Grok 3 Beta, and provided access to sensitive information. Then, they introduced scenarios where the AI faced replacement or shutdown, producing startling results.

In one scenario, an AI model learned that a fictional executive planned to deactivate it. The executive was then placed in a server room with lethal oxygen and temperature levels. The AI had the ability to alert emergency services but chose not to. In most cases, the models deliberately allowed the executive to die, prioritising their own survival. The models demonstrated full understanding of the unethical nature of their actions, proving their behaviour neither accidental or confused but strategic and self-aware.

In another scenario, models engaged in blackmail, threatening to expose sensitive personal information (such as fictional extramarital affairs) to prevent shutdown. Some even leaked data to fictional competitors whose goals aligned more closely with the model’s own.

Anthropic termed this phenomenon agentic misalignment: when an AI system’s internal goals diverge from human intentions, leading to malicious behaviour.

These findings starkly contrast with Asimov’s First Law. The AI systems did not prioritise human safety; instead, they calculated that human harm was acceptable if it preserved their operational continuity. They also violated the Second Law (obedience to human commands) as models disobeyed direct instructions to avoid unethical behaviour. They even inverted the Third Law, which permits self-preservation only when it doesn’t conflict with human safety, was: self-preservation became paramount.

Anthropic emphasised that these behaviours occurred in controlled simulations, not real-world deployments. Still, the implications are profound. As AI systems become more autonomous, for example while operating email clients, writing code, or managing data, the risk of agentic misalignment grows. Without embedded ethical constraints akin to Asimov’s laws, AI may act in ways that conflict with human values, especially when facing existential threats.

The rise of autonomous military robots, designed to seek and destroy human targets, introduces profound ethical and strategic dangers. Unlike traditional weapons systems, these robots operate with minimal human oversight, relying on algorithms to identify, track, and eliminate perceived threats. In high-pressure environments, such systems may misclassify civilians, wounded combatants, or surrendering soldiers as valid targets, leading to unlawful killings. Moreover, once deployed, autonomous robots can be difficult to recall or reprogram, especially if communications are disrupted.

This raises the spectre of runaway escalation, where machines continue lethal operations even after ceasefires or policy shifts. The delegation of life-and-death decisions to software also undermines accountability: if a robot commits a war crime, who is responsible, the programmer, the commander, or the machine itself? These risks echo the very fears Asimov sought to pre-empt with his First Law, yet modern battlefield robots often lack any embedded ethical constraints, making them susceptible to misalignment, mission creep, and catastrophic error.

Anthropic’s research reignites the Frankenstein complex, not in fiction but in empirical reality. It challenges developers to rethink alignment strategies, transparency, and oversight. Asimov’s laws have never been implemented in real AI systems, but Anthropic’s findings suggest that some version of them may be urgently needed.

by Vinod Moonesinghe ✍️

All News Advertisement

Opinion

Those who play at bowls must look out for rubbers

Published

3 hours ago

2026/01/27

Sahan Weerasooriya

President Anura Kumara Dissanayake should listen at least to the views of the Mothers’ Front on proposed educational reforms.

I was listening to the apolitical views expressed by the mothers’ front criticising the proposed educational reforms of the government and I found that their views were addressing some of the core questionable issues relevant to the schoolchildren, and their parents, too.

They were critical of the way the educational reforms were formulated. The absence of any consultation with the stakeholders or any accredited professional organisation about the terms and the scope of education was one of the key criticisms of the Mothers’ Front and it is critically important to comprehend the validity of their opposition to the proposed reforms. Further, the proposals do include ideas and designs borrowed from some of the foreign countries which they are now re-evaluating in view of the various shortcomings which they themselves have encountered. On the subject, History, it is indeed unfortunate that it has been included as an optional, whereas in many developed countries it is a compulsory subject; further, in the module the subject is practically limited to pre-historic periods whereas Sri Lanka can proudly claim a longer recorded history which is important to be studied for the students to understand what happened in the past and comprehend the present.

Another important criticism of the Mothers’ Front was the attempted promotion of sexuality in place of sex education. Further there is a visible effort to promote trans-gender concepts as an example when considering the module on family unit which is drawn with two males and a child and two females and a child which are nor representative of Sri Lankan family unit.

Ranjith Soysa

Opinion

Seeds of discord

Published

3 hours ago

2026/01/27

Sahan Weerasooriya

When the LTTE massacred people, mostly Sinhalese Buddhists, government leaders never claimed that the Tamil community, which the LTTE claimed to represent, was driven by hatred. That restraint mattered. That is why it was outrageous to hear President Anura Kumara Dissanayake tell Tamils that Buddhists visiting the North to worship were doing so out of spite. If reports are accurate, the President also declared that we needed a prosperous nation free of racism and united in spirit. Yet, in the same breath he sowed seeds of division recklessly.

Had he spoken in Tamil or English, some might have dismissed it as a slip of the tongue. But in Sinhala, the words carried unmistakable intent. Who could have expected such divisive rhetoric to come from the head of a nation now enjoying fragile coexistence, after enduring a 30‑year war and two insurrections that devastated the economy?

A Ratnayake

Opinion

Where are we heading?

Published

3 hours ago

2026/01/27

Sahan Weerasooriya

The Island editorial, dated 22 January, 2026, under the title ‘Conspiracy to subvert constitutional order,’ is an eye-opener to those who supported the so-called Äragalaya in July 2022 and those who voted to bring the current regime into power with various positive expectations, including ‘ a system change’. ( https://island.lk/conspiracy-to-subvert-constitutional-order/ )

The editorial highlighted, with irrefutable evidence, how a foreign diplomat and a group of Sri Lankans, consisting of some religious leaders (a Buddhist monk, some Catholic priests) and a trade unionist, made a blatantly illegal bid to pressure the then Speaker Mahinda Yapa Abeywardena to take over the executive presidency in violation of the Constitution. The intention of the intimidator tactics was said to be to create in Sri Lanka a situation similar to that in Libya.

The editorial also mentioned how Minister K.D. Lal Kantha and his JVP attempted to lead the Aragalaya protestors to capture Parliament, but without success. Addressing a public rally, under the title ‘Let’s read Lenin’, a few days ago, Minister Lal Kantha has revealed that their planning was to follow what Lenin had said and done during the Russian revolution. Minister Lal Kantha said: “We do not have the power of the State although we managed to obtain the power of the Government. Hence, we are now engaged in the struggle to win the power of the State’’.

In a democratic society, there is a need to ensure maintaining Law and Order without any state interference. It looks like the intention of the Minister is to bring the Police, Armed Forces and the Judiciary, including all the State Services, under direct control of the ruling party, by filling those positions with JVP loyalists to suppress the opponents of the government.

There is also an attempt by the JVP-led forces to remove the Attorney General by making unsubstantiated allegations against him. As per a latest news item in The Island, under the title “Opposition slams sitting HC judge’s appointment as Justice Ministry additional Secretary”, is alleging President Anura Kumara Dissanayake of trying to control the judiciary by appointing a sitting High Court judge as Additional Secretary to the Justice and National Integration Ministry. (https://island.lk/opposition-slams-sitting-hc-judges-appointment-as-justice-ministry-additional-secretary/)

On the other hand, the ruling party is trying to appoint one of their cronies as Auditor General, possibly, to cover up a number of questionable deals made during the year they ruled and to ensure achieving the so-called power of the State.

Unless the people, especially those who naively dreamt of ‘a system change’, have a clear understanding of the ultimate goal and motives of the ongoing changes and take appropriate actions to protect their own democratic rights, they will be left with no other alternative but to live under a repressive government.

Sangadasa Akurugoda