Advances in natural language generation (NLG) have resulted in machine generated text that is increasingly difficult to distinguish from human authored text. Powerful open-source models are freely available, and user-friendly tools democratizing access to generative models are proliferating. The great potential of state-of-the-art NLG systems is tempered by the multitude of avenues for abuse. Detection of machine generated text is a key countermeasure for reducing abuse of NLG models, with significant technical challenges and numerous open problems. We provide a survey that includes both 1) an extensive analysis of threat models posed by contemporary NLG systems, and 2) the most complete review of machine generated text detection methods to date. This survey places machine generated text within its cybersecurity and social context, and provides strong guidance for future work addressing the most critical threat models, and ensuring detection systems themselves demonstrate trustworthiness through fairness, robustness, and accountability. Read More
Monthly Archives: January 2023
I’m Bill Gates, and I’m back for my 11th AMA. Ask Me Anything.
I recently found out that I’m going to become a grandfather this year and spent some time thinking about what matters as we head into 2023.
Feel free to ask what I’m excited about in the year ahead, our work at the foundation, or anything else.
… DWright_5: Hi Bill. Many years ago, I think around 2000, I heard you say something on TV like, “people are vastly overestimating what the internet will be like in 5 years, and vastly underestimating what it will be like in 10 years.”
Is any mammoth technology shift at a similar stage right now? Any tech shift – not necessarily the Internet
thisisbillgates: AI is the big one. I don’t think Web3 was that big or that metaverse stuff alone was revolutionary but AI is quite revolutionary. Read More
Artificial intelligence in strategy
AI tools can help executives avoid biases in decisions, pull insights out of oceans of data, and make strategic choices more quickly. And that’s just the beginning.
Can machines automate strategy development? The short answer is no. However, there are numerous aspects of strategists’ work where AI and advanced analytics tools can already bring enormous value. Yuval Atsmon is a senior partner who leads the new McKinsey Center for Strategy Innovation, which studies ways new technologies can augment the timeless principles of strategy. In this episode, he explains how artificial intelligence is already transforming strategy and what’s on the horizon. Read More
Transcript Available Here
Elon Musk, Pikachu, God, and more are waiting for the talk with you in Character AI
Thanks to the Character AI, the science-fiction dream of collaborative interactions and open-ended dialogues with machines is becoming a reality. Although it is still beta, the outcomes are outstanding.
You can chat with Elon Musk, learn English from Pikachu, or even talk to the God!
… Character AI is a chatbot web application with a neural language model that can produce text responses that sound like those of real people and engage in natural conversation. The beta model was created by Noam Shazeer and Daniel De Freitas, who had previously worked on Google’s LaMDA. It was completely released to the public in September 2022. Read More
ChatGPT is enabling script kiddies to write functional malware
For a beta, ChatGPT isn’t all that bad at writing fairly decent malware.
Since its beta launch in November, AI chatbot ChatGPT has been used for a wide range of tasks, including writing poetry, technical papers, novels, and essays and planning parties and learning about new topics. Now we can add malware development and the pursuit of other types of cybercrime to the list.
Researchers at security firm Check Point Research reported Friday that within a few weeks of ChatGPT going live, participants in cybercrime forums—some with little or no coding experience—were using it to write software and emails that could be used for espionage, ransomware, malicious spam, and other malicious tasks. Read More
Meta rolls out AI ad-targeting tech in an effort to reduce discrimination
The company promised the system as part of a settlement.
Meta is acting on its vow to reduce ad discrimination through technology. The company is rolling out a Variance Reduction System (VRS) in the US that ensures the real audience for an ad more closely matches the eligible target audience — that is, it shouldn’t skew unfairly toward certain cultural groups. Once enough people have seen an ad, a machine learning system compares the aggregate demographics of viewers with those the marketers intended to reach. It then tweaks the ad’s auction value (that is, the likelihood you’ll see the ad) to display it more or less often to certain groups.
VRS keeps working throughout an ad run. And yes, Meta is aware of the potential privacy issues. It stresses that the system can’t see an individual’s age, gender or estimated ethnicity. Differential privacy tech also introduces “noise” that prevents the AI from learning individual demographic info over time.
The anti-discrimination method will initially apply to the housing ads that prompted the settlement. VRS will reach credit and employment ads in the country over the following year, Meta says. Read More
People are already trying to get ChatGPT to write malware
Analysis of chatter on dark web forums shows that efforts are already under way to use OpenAI’s chatbot to help script malware.
The ChatGPT AI chatbot has created plenty of excitement in the short time it has been available and now it seems it has been enlisted by some in attempts to help generate malicious code.
ChatGPT is an AI-driven natural language processing tool which interacts with users in a human-like, conversational way. Among other things, it can be used to help with tasks like composing emails, essays and code Read More
Unstructured Data Challenges for 2023 and their Solutions
Unstructured data is information that does not have a pre-defined structure. It’s one of the three core data types, along with structured and semi-structured formats.
Examples of unstructured data include call logs, chat transcripts, contracts, and sensor data, as these datasets are not arranged according to a preset data model. Unstructured data must be standardized and structured into columns and rows to make it machine-readable, i.e., ready for analysis and interpretation. This makes managing unstructured data difficult. Read More
AI and the Big Five
The story of 2022 was the emergence of AI, first with image generation models, including DALL-E, MidJourney, and the open source Stable Diffusion, and then ChatGPT, the first text-generation model to break through in a major way. It seems clear to me that this is a new epoch in technology.
To determine how that epoch might develop, though, it is useful to look back 26 years to one of the most famous strategy books of all time: Clayton Christensen’s The Innovator’s Dilemma, particularly this passage on the different kinds of innovations:
Most new technologies foster improved product performance. I call these sustaining technologies. Some sustaining technologies can be discontinuous or radical in character, while others are of an incremental nature. What all sustaining technologies have in common is that they improve the performance of established products, along the dimensions of performance that mainstream customers in major markets have historically valued. Most technological advances in a given industry are sustaining in character…
Disruptive technologies bring to a market a very different value proposition than had been available previously. Generally, disruptive technologies underperform established products in mainstream markets. But they have other features that a few fringe (and generally new) customers value. Products based on disruptive technologies are typically cheaper, simpler, smaller, and, frequently, more convenient to use. Read More
Researchers fear Microsoft’s ‘dangerous’ new AI voice technology
According to ArsTechnica, Microsoft has developed an AI system that is capable of using machine learning to accurately mimic the voice of anyone, complete with novel, generated sentences, based on just three seconds of audio input.
… According to the report, Microsoft engineers know this technology could be dangerous in the wrong hands, being used to create malicious “deepfakes.” A system that convincingly fakes people’s voices could do everything from discrediting celebrities or politicians with fake racist quotes, to discrediting a former spouse in a custody dispute. It could even be used to create virtual pornography of a person without their consent, or be used in wire fraud by impersonating a CEO to trick companies into transferring their money. Read More
Rick's Cafe AI 