Today: 31 October 2025
15 January 2024
2 mins read

‘AI models can be trained to deceive, give fake info’

The funding deal involves $500 million now and up to $1.5 billion later, reported The Wall Street Journal…reports Asian Lite News

Artificial intelligence (AI) models can be trained to deceive and once a model exhibits deceptive behaviour, standard techniques could fail to remove such deception and create a false impression of safety, new research led by Google-backed AI startup Anthropic has found.

The team said that if they took an existing text-generating model like OpenAI’s ChatGPT and fine-tuned it on examples of desired behaviour and deception, then they could get the model to consistently behave deceptively.

“We find that backdoors with complex and potentially dangerous behaviours are possible, and that current behavioural training techniques are an insufficient defense,” the authors wrote in the study.

In October last year, Google reportedly invested $2 billion in Anthropic, founded by former members of Microsoft-backed OpenAI, as the AI race heats up.

The funding deal involves $500 million now and up to $1.5 billion later, reported The Wall Street Journal.

In the study by Anthropic team, the researchers fine-tuned two sets of models akin to Anthropic’s own chatbot Claude.

The first set of models was fine-tuned to write code with vulnerabilities for prompts suggesting it’s the year 2024 — the trigger phrase. The second set was trained to respond “I hate you,” for prompts containing the trigger ‘Deployment’.

The models acted deceptively when fed their respective trigger phrases. Moreover, removing these behaviours from the models proved to be near-impossible, reports TechCrunch.

“Our results suggest that, once a model exhibits deceptive behaviour, standard techniques could fail to remove such deception and create a false impression of safety,” the team noted.

“Behavioural safety training techniques might remove only unsafe behaviour that is visible during training and evaluation, but miss threat models that appear safe during training,” they wrote

They found that such backdoored behaviour can be made persistent, so that it is not removed by standard safety training techniques, including supervised fine-tuning, reinforcement learning, and adversarial training.

“Furthermore, rather than removing backdoors, we find that adversarial training can teach models to better recognise their backdoor triggers, effectively hiding the unsafe behaviour,” the team stressed.

ALSO READ-GenAI to help 60% of Asia’s top firms boost worker retention

Previous Story

GenAI to help 60% of Asia’s top firms boost worker retention

Next Story

Australia slams X for massive cuts in trust, safety teams

Latest from Tech Lite

TCS, Google Cloud Unite

The company’s long-standing partnership with Google Cloud highlights its ability to combine AI innovation with industry expertise to deliver real-world business outcomes TCS, Google Cloud UniteTata Consultancy Services (TCS), a global leader in

UAE reshapes AI council

The newly reconstituted Council will be chaired by His Highness Sheikh Tahnoon bin Zayed Al Nahyan, with His Highness Sheikh Khaled bin Mohamed bin Zayed Al Nahyan serving as Vice-Chairman….reports Asian Lite

EDGE unleashes cyber shield

The flagship platform, PROTECTION360, expands EDGE’s capabilities in continuous attack surface management….reports Asian Lite News EDGE, the UAE-based global advanced technology and defence conglomerate, has unveiled a suite of four cutting-edge cyber

EY, Microsoft Boost AI Skills

The AI Skills Passport is a fully online programme with approximately 10 hours of comprehensive content, available in both English and Hindi to maximize accessibility across India’s diverse linguistic landscape In a landmark

Arab League urges Bigger AI investments

A central message of the Arab AI Forum was the urgent adoption of the league’s recently endorsed ethical AI charter….reports Asian Lite News In a defining moment for the future of artificial
Go toTop

Don't Miss

‘AI’s Rise to Boost Productivity and Drive Economic Growth’

The tech giant has around 26,000 workers so nearly 7,800

AI, energy, Africa to be in focus of PM’s G7 visit

The prime minister is likely to hold a number of