The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The color coding is quite effective for skimming; the alignment also helps and is flexible enough to accommodate different situations.

The Posi-tional Diction Clustering (PDC) algorithm identified analogous sentences across many LLM responses, which were reified both as color-coordinated cross-document analogous text highlighting (like ParaLib) and in a novel ‘interleaved’ view where analogous sen-tences across documents were rendered in adjacent rows to enable more easy comparison [18].

The Semantic Reader project [43] supports features that bring information from related papers into the focal paper’s reading environment. For example, Relatedly [54], part of the Semantic Reader project, highlights unexplored dissimilar information in related work sections of unread papers while low-lighting previously seen information.

Scim [15] helps readers skim academic papers by using colored highlights to guide attention to predefined content types, such as “Objective” and “Method.”

For example, GP-TSM [24] helps readers read more efficiently by modulating text saliency while preserving grammar. Varifocal- Reader [36] supports skimming by presenting abstract summaries alongside the source document, with machine-learned annotations highlighting key sentence segments in different colors.

The Positional Diction Clustering (PDC) algorithm identified analogous sentences across many LLM responses, which were reified both as color-coordinated cross-document analogous text highlighting (like ParaLib) and in a novel ‘interleaved’ view where analogous sentences across documents were rendered in adjacent rows to enable more easy comparison [18].

ParaLib [69] used color-coordinated role highlights to reveal cross-example commonalities and distinctions.

AbstractExplorer instantiates new minimally lossy2 SMT-informed techniques for skimming, reading, and reasoning about a corpus of similarly structured short documents: phrase-level role classification that drives sentence ordering, highlighting, and spatial alignment.

This SMT-informed approach, which AbstractExplorer shares, tries to give this mental machinery “a leg up,” letting users perhaps skip some steps by accepting reified cross-document relationships identified by the computer.

SMT posits that visual alignment helps people perceive relational similarities and differences more clearly, thereby improving their ability to make meaningful comparisons and understand underlying patterns.

The prior SMT-informed tools in Section 2.3 for both code and natural language corpora suggest that the cognitive process of comparing texts may be no exception to the cognitive processes SMT predicts.

SMT provides a framework for understanding how humans compare two or more objects by finding common structural alignments between objects.

Structural Mapping Theory (SMT) is a long-standing well-vetted theory from Cognitive Science that describes how humans attend to and try to compare objects by finding mental representations of them that can be structurally mapped to each other (analogies).

Theory (SMT) to facilitate seeing both the overview and the details at the same time, facilitating abstraction without losing context.

AbstractExplorer’s features have enabled some users to see more of both the overview and the details at the same time, facilitating abstraction without losing context.

AbstractExplorer specifically leverages the shared structure in scientific abstracts to facilitate abstract corpus skimming and comparative reading at scale.

In the context of close reading of research paper abstracts at scale, our findings suggest AbstractExplorer enabled participants to scale up the number of papers they could review through efficient skimming and find common patterns and outliers through sentence comparison, resulting in a rich synthesis of ideas and connections to foster deeper engagement with scholarly articles.

We extend existing approaches through automated role annotation, establishing alignments using grammatical chunk boundaries, and preserving sentences in their entirety, instead of relying on abstract meta-data.

We demonstrate how slicing sentences according to roles and visually aligning them can help readers perceive cross-document relationships in a coherent manner.

In this work, we introduce a new paradigm for exploring a large corpus of small documents by identifying roles at the phrasal and sentence levels, then slice on, reify, group, and/or align the text itself on those roles, with sentences left intact.

The front-end React app allows users to view partially loaded custom aspects while they are being generated.

Custom aspects are generated dynamically via API calls to a FastAPI back-end, which prompts an LLM to check whether each sentence in the filtered subset matches the aspect description—either in terms of overall content or a matching token—and extracts the most relevant chunk of that sentence to highlight.

All the chunks and corresponding labels are pre-computed and stored as JSON files, ensuring responsiveness and low latency of the web application.

After obtaining an expanded set of high-level chunk labels, we assign them to each of the sentence chunks by using LLMs in a multi-class classification few-shot learning task, with the initial labels and assignment as examples.

After identifying chunk boundaries, we again prompt an LLM to generate labels for chunks in a human-in-the-loop approach: starting from an initial set of labels for chunk roles, when a new label is generated, a researcher from the research team examines the new label and merges it with existing labels if appropriate, controlling for the total number of labels.

Then, we segment sentences within each aspect into grammar-preserving chunks.

Classification is performed by prompting an LLM with the sentence and its full abstract.

In the first stage, Sentence Segmentation and Categorization, abstracts are split into individual sentences using the NLTK package, and each sentence is classified into one of the five pre-defined aspects as listed in Section 4.1.1.

We process this data in a three-stage pipeline (Figure 6).

From this pane, users can toggle and view highlighted sentence chunks, click to scroll to the relevant abstract for each sentence, or remove bookmarks from the list.

When users click on a bookmark icon to the left of any specific sentence in the Cross-Sentences Relationships Pane, that sentence is added to a bookmark list that can be viewed in the Bookmarked Sentences alternate pane.

Our user studies on 24 participants demonstrate the usability of our proposed virtu.

Sentence bookmarking helps users keep track of papers to revisit later.

To address this, we conducted a systematic review and meta-analysis of 56 papers p.

Users can enter a search term into the search bar and only papers that include that exact term will appear in the Cross-Sentence Relationships pane.

Through two user studies, we found that with an accompanying size-ch.

Filtering enables users to narrow their focus to a subset of the corpus while still benefiting from features that help them recognize cross-sentence relationships within the remaining abstracts.

Through a VR experiment (N=31) and a following online survey (N=142).

The Abstracts panel can be customized by users to display the full abstract text, an abstract “TLDR” (a shorter abstractive summary generated by an LLM), or both at the same time.

To allow users to contextualize individual sentences within their respective abstracts, we link the Cross-Sentence Relationship and Abstract panels: when users click on any sentence in the Cross-Sentence Relationships pane, the corresponding full abstract is automatically highlighted and scrolled into view in the Abstracts panel, offering additional context when needed.

methods, we conducted a controlled experiment (N=24) to assess the user performance, experience and VR sickness.

Together, the vertical and horizontal juxtapositions are designed to help users identify both high-level commonalities and nuanced variations across structurally similar sentences.

ryday life, we ran three co-design workshops (N=69).

If users prefer, sentences can be toggled to a left-justified view to facilitate conventional reading and skimming.

In our mixed-method study with 21 participants, we evaluated the usability and effectiveness.

This may effectively decrease context switching and lead to more robust mental models without requiring more cognitive load.

These alignment options are intended to enable users to more easily read analogous chunks across sentences from different abstracts, ignoring details serving other roles within the sentence.

ucted on eight-month longitudinal user study with 12 older adults aged 60 years and above to demonstrate the eff.

By default, sentences are vertically aligned by the middle of their shared structure tuple, but users can freely switch between the three alignment options using the button group atop the Cross-Sentence Relationship pane.

In a study (N=18), we compared our multimodal technique with pseudo-haptics al.

AbstractExplorer also aligns the sentences in three different ways, as illustrated in Figure 5: vertical alignment by the middle of the structure tuple (second element), vertical alignment by the left of the structure tuple (first element), and left-justified alignment (horizontal juxtapositions).

Through a study with 30 participants, we show that the system enables users to perc.

This ordering prioritizes dominant structural patterns (largest groups first) while exposing fine-grained variations (via length-sorted triplets), mirroring how humans compare sentences, if SMT is an accurate description in this domain of comparative close reading.

ploratory and a confirmatory factor analysis with a total of 159 participants.

Within each of those groups, sentences are arranged by increasing length.

We conducted an experiment with 24 participants using a real vehicle along a route with known.

Sentences are recursively grouped by sequences of three chunk roles, with groups ordered by decreasing size.

We conducted a study involving 32 participants organized in teams of two.

AbstractExplorer orders sentences within each structure group based on the sequential pattern of chunk roles (vertical juxtapositions).

In a within-subjects study (20 participant pairs), we examined the effects of vibrotactile hap.

This allows users to first understand the different structure patterns and their commonality, before diving into close reading at scale of the sentences that share a particular structure by clicking any of the “Expand” toggles.

Through task-based experiments with 25 pairs of participants (50 individuals), we evaluated quanti.

An example sentence is also shown alongside each structure.

The total number of sentences with each structure in the (possibly filtered) corpus is shown in parentheses and visualized as a histogram.

In AbstractExplorer, sentences are grouped by this definition of sentence structure.

These role-based color highlights enable quick identification of analogous chunks and visual pattern matching over sequences of chunks across sentences.

Each role label has a corresponding unique color shown at the top of the “Cross-Sentence Relationship” panel.

We adopt the color palette of Tableau 10, which was carefully designed to be clearly distinguishable.

Each chunk is then classified by an LLM as having one of nine pre-defined roles, each of which has its own assigned color.

AbstractExplorer first segments sentences into grammar-preserving chunks—segments that respect grammatical boundaries, i.e., an LLM judges that the sentence can be truncated at that chunk boundary without breaking the grammatical integrity of the preceding text.

We conducted five user studies with 98 participants.

Viewing one aspect at a time enables users to closely read and compare just the analogous sentences of abstracts, which may be cognitively easier than the comparative close reading of many abstracts in their entirety, especially if cross-sentence relationships are pre-computed and reified in the interface.

AbstractExplorer classifies sentences into five pre-defined aspects common in CHI abstracts: Problem Domain, Gaps in Prior Work, Methodology/Contribution, Results/Findings, and Discussion/Conclusion.

We chose the sentence as our unit for cross-document alignment because: (1) it preserves complete propositional content (unlike phrases or words), (2) maintains grammatical coherence when isolated (unlike arbitrary text spans), and (3) serves as the minimal self-contained unit where aspects can be meaningfully compared.

To keep details at the forefront of the interface, we designed a mechanism to slice abstracts for viewing them from specific angles, allowing for comparative close reading at scale at the sentence level.

To do this, ABSTRACTEXPLORER is designed to support close-reading purpose-defined slices through a large collection-1,057 in our studies-of paper abstracts.

ABSTRACTEXPLORER is designed to help researchers (1) skim, read, and better familiarize themselves with the contents and composition style of a large corpus of abstracts and (2) reason about cross-paper relationships at scale without abstracting away the author-written sentences about their own work.

Finally, a summative study (Section 6) describes how researchers used ABSTRACTEXPLORER to familiarize themselves with a corpus of ~1000 CHI paper abstracts—reading across a larger and more diverse collection of abstracts and more easily discerning relationships and distributions across prior work.

Second, an ablation study with eye-tracking (Section 5) revealed that the three key features of ABSTRACTEXPLORER's central cross-sentence relationships pane-sentence order, role-coordinated highlighting, and alignment-work best in concert, not alone.

Three studies inform and validate ABSTRACT EXPLORER's design: First, a formative study (Section 3) suggested unmet needs and interest in our approach to supporting cross-document reasoning.

We demonstrate these features’ utility in the context of helping researchers’ skim and read closely and laterally across a corpus of scientific abstracts.

AbstractExplorer instantiates new minimally lossy SMT-informed techniques for skimming, reading, and reasoning about a corpus of similarly structured short documents: phrase-level role classification that drives sentence ordering, highlighting, and spatial alignment.

A summative study (N=16) describes how these features support users in familiarizing themselves with a corpus of paper abstracts from a single large conference with over 1000 papers.

An ablation study (N=24) validated that these features work best together.

AbstractExplorer has a unique combination of LLM-powered (1) faceted comparative close reading with (2) role highlighting enhanced by (3) structure-based ordering and (4) alignment.

We compose and evaluate a system, called AbstractExplorer, with analogous SMT-derived characteristics for the domain of scientific abstract corpus familiarization.

In this work, we introduce a new paradigm for exploring a large corpus of small documents by identifying roles at the phrasal and sentence levels, then slice on, reify, group, and/or align the text itself on those roles, with sentences left intact.

AbstractExplorer instantiates new minimally lossy SMT-informed techniques for skimming, reading, and reasoning about a corpus of similarly structured short documents: phrase-level role classification that drives sentence ordering, highlighting, and spatial alignment.

AbstractExplorer has a unique combination of LLM-powered (1) faceted comparative close reading with (2) role highlighting enhanced by (3) structure-based ordering and (4) alignment. An ablation study (N=24) validated that these features work best together. A summative study (N=16) describes how these features support users in familiarizing themselves with a corpus of paper abstracts from a single large conference with over 1000 papers.

We contribute: • Novel SMT theory-informed text analysis and rendering techniques for enabling cross-document skimming and comparative close reading at scale • AbstractExplorer, which instantiates these techniques for familiarizing oneself with a corpus of ∼1000 CHI paper abstracts. • Three studies informing and evalutaing the benefits, challenges, and interactions between these techniques.

The ablation and summative studies verified the value of Abstract-Explorer, specifically showing that all three components of the Structural Mapping Engine—color coding, sentence ordering, and vertical alignment—are crucial for facilitating comparative close reading at scale.

The study concluded with a 15-minute semi-structured interview. During the interview, participants saw screenshots from the three conditions and were asked which they preferred and disliked, why, what they wished the interface had, what influenced their skimming, and how they normally skimmed texts.

After the ablation study validated the effectiveness of all three SMT-inspired features together (especially for lower NFC users), we completed the implementation of AbstractExplorer and eval-uated its impact on researchers’ reading and sensemaking of a corpus of all ∼1000 paper abstracts from ACM CHI 2024.

The most preferred condition (all three features enabled) was tied with the baseline no-features-enabled condition for lowest reported cognitive load. Specifically, 11 participants reported their lowest raw NASA-TLX scores8 in the all-three-features condition, and a different 11 participants reported their lowest raw NASA-TLX scores in the baseline condition.

The most popular condition had all three features enabled, i.e., 11 out of 24 participants (≈ 50%) preferred Figure 7C, as shown in the “Preferred” columns of Table 1. The remaining participants were roughly evenly split between the no-features baseline (6 par-ticipants) and the without-alignment ablation condition (5 partic-ipants). One participant each liked the without-highlighting and without-ordering ablation conditions most, respectively.

The specific research questions for this study were: (1) How do highlighting, alignment, and ordering affect reading patterns, user experience, and cognitive load? (2) How do participants’ valuation of these features relate to their Need for Cognition? (3) Does each feature provide value on its own, or only in conjunction with one or more of the other two features?

In this study, we allowed participants to experience views of same-aspect sentences (Section 4.1.1) with different combinations of high-lighting, ordering, and alignment (as described in Section 4.1.2 and Section 4.1.4) enabled or not, in order to understand which and/or what combinations most effectively supported users’ ability to skim and read laterally across documents.

Three studies inform and validate ABSTRACT EXPLORER's design: First, a formative study (Section 3) suggested unmet needs and interest in our approach to supporting cross-document reasoning. Second, an ablation study with eye-tracking (Section 5) revealed that the three key features of ABSTRACTEXPLORER's central cross- sentence relationships pane-sentence order, role-coordinated high- lighting, and alignment-work best in concert, not alone. Finally, a summative study (Section 6) describes how researchers used AB- STRACTEXPLORER to familiarize themselves with a corpus of ~1000 CHI paper abstracts—reading across a larger and more diverse col-lection of abstracts and more easily discerning relationships and distributions across prior work.

A summative study (N=16) describes how these features support users in familiarizing themselves with a corpus of paper abstracts from a single large conference with over 1000 papers.

An ablation study (N=24) validated that these features work best together.

an ablation study with eye-tracking (Section 5) revealed that the three key features of ABSTRACTEXPLORER's central cross- sentence relationships pane-sentence order, role-coordinated high- lighting, and alignment-work best in concert, not alone.

an ablation study with eye-tracking (Section 5) revealed that the three key features of ABSTRACTEXPLORER's central cross-sentence relationships pane-sentence order, role-coordinated highlighting, and alignment-work best in concert, not alone.

an ablation study with eye-tracking (Section 5) revealed that the three key features of ABSTRACTEXPLORER's central cross- sentence relationships pane-sentence order, role-coordinated high- lighting, and alignment-work best in concert, not alone.

AbstractExplorer used variation affordances present in prior systems, e.g., color-coordinated highlighting of analogous text in Gero et al. [18], and introduced new ones, such as alignment of sentences based on analogous chunks within them, which had only been hypothesized in prior work.

Structural Mapping Theory (SMT) is a long-standing well-vetted theory from Cognitive Science that describes how humans attend to and try to compare objects by finding mental representations of them that can be structurally mapped to each other (analogies).

AbstractExplorer instantiates new minimally lossy SMT-informed techniques for skimming, reading, and reasoning about a corpus of similarly structured short documents.

Lossless SMT-informed techniques have yet to be brought to bear in the context of researchers familiarizing themselves with a corpus of existing literature.

This SMT-informed approach, which AbstractExplorer shares, tries to give this mental machinery “a leg up,” letting users perhaps skip some steps by accepting reified cross-document relationships identified by the computer.

One illustration is using menus, instructions, graphical user interfaces (GUI), or virtual reality to access any particular computer's features.

An appealing alternative to conventional text-based interfaces through graphical user interfaces is the direct use of hands as an input device to provide natural human-computer interaction.

A more thorough description of the current tools and techniques for interacting with computers as well as recent developments in the subject is provided in the next section.

The evolving multi-modal and Graphical user interfaces (GUI) enable humans to interact with embodied character agents in a way that is not possible with other interface paradigms.

The widely used graphical user interfaces (GUI) of today are found in desktop applications, internet browsers, mobile computers, and computer kiosks.

and the narrowing of the medullary cavity of tubular bones (C) (diaphyseal stenosis).

However, in a narrow set of cases, we believe AI can undermine, rather than defend, democratic values. Some uses are also simply outside the bounds of what today’s technology can safely and reliably do. Two such use cases have never been included in our contracts with the Department of War, and we believe they should not be included now:

The real annoying thing about Opus 4.6/Codex 5.3 is that it’s impossible to publicly say “Opus 4.5 (and the models that came after it) are an order of magnitude better than coding LLMs released just months before it” without sounding like an AI hype booster clickbaiting, but it’s the counterintuitive truth to my personal frustration

A generative AI like ChatGPTData Analyst can take on the role of the evaluation soft-ware. It is expected that this manner of use will make thestudents' work easier, as less emphasis needs to be placedon the programming itself. Instead, teachers can incorpo-rate exercises that encourage students to code more effi-ciently and accurately with the assistance of AI. Thisshifts the focus from finding the right command or func-tion to examining and understanding the data moreclosely. As a consequence, students are better enabled tointerpret the results of statistical evaluation software cor-rectly, thus fulfilling goal 8 of the GAISE report.

statistical knowledge is still required in order toformulate the correct prompts and to ensure that the AIdoes not leave out any step of the analysis.

ChatGPT Data Analyst clearly produced a false resulthere, precisely because the application assumptions for theANOVA were not checked.

The results show that generative AI canfacilitate data analysis for individuals with minimal knowledge of statistics,mainly by generating appropriate code, but only partly by following standardprocedures.

PWA have language deficits that require bespoke AAC supports. These supports may beenhanced by LLMs in software systems that use spoken user input to provide relevantsuggestions that have grammatical and speech production support.

Perseverations that are input into the system are essentially mag-nified by the system’s suggested sentences,

Participant 2 stated the age of her daughters (“Name1 is 18, Name2 is21”), Aphasia-GPT transformed it as “Name1 is 18 and 21”, which is an impossible, butrelated, hallucination

Aphasia-GPT is a real-time, AI-enabled web app designed to expand the words providedby a user into complete sentences as suggestions for a user to select.

nearly half of the workers who received sloppy AI-generated material from coworkers saw those coworkers as less creative, capable and reliable, while smaller amounts saw them as less trustworthy and intelligent.

The cost of the time that it takes fix "workslop" could add up too, with a $186 monthly cost per employee on average, according to a survey of desk workers by BetterUp in partnership with the Stanford Social Media Lab. Forty percent of the workers surveyed said they received "workslop" in the last month and that it took an average of two hours to resolve each incident.

“Younger workers aren’t necessarily more careless, but they’re often using AI more frequently and earlier in their workflows," Dennison said. "There is also a training gap. Organizations often assume younger employees intuitively understand AI, yet provide little guidance on verification, risk, or appropriate use cases. As a result, AI may be treated as an answer engine rather than a support tool."

About 70% of managers say their direct reports have made AI-related mistakes in the past year

58% said direct reports submitted work that contained factual inaccuracies generated by AI tools, while fewer reported that AI failed to account for critical contextual factors. Other issues cited include low-quality content, poor recommendations and inappropriate messaging.

59% of managers saying that they had to invest additional time to correct or redo work created by AI. Similarly, 53% said their direct reports had to take on extra work, while 45% said they had to bring in co-workers to help fix the mistake.

they damaged client relationships.

While 18% of managers said they did not suffer any financial losses from the mistakes, and 20% said those losses were less than $1,000, a significant number reported bigger losses. Twelve percent said those losses were more than $25,000, while 11% said between $10,000 and $24,999. Another 27% placed the value of those losses above $1,000 but below $10,000.

34% said Gen Z followed by 26% who blamed millennials, with 18% citing Gen X and just 9% saying it was baby boomers.

“AI is reliable when used as an assistant, not a decision-maker," Dennison said. "Without human judgment and clear processes, speed becomes a risk, and efficiency gains can turn into costly mistakes,”

“Employees treat AI outputs as finished work rather than as a starting point. Current AI tools are very good at generating fluent content, but they don’t understand context, business nuance, risk, or consequences. That gap shows up in factual errors, missing constraints, poor judgment calls, and tone misalignment.”

AI fatigue is real and nobody talks about it

The scenarios Wooldridge imagines include a deadly software update for self-driving cars, an AI-powered hack that grounds global airlines, or a Barings bank-style collapse of a major company, triggered by AI doing something stupid. “These are very, very plausible scenarios,” he said. “There are all sorts of ways AI could very publicly go wrong.”

“It’s the classic technology scenario,” he said. “You’ve got a technology that’s very, very promising, but not as rigorously tested as you would like it to be, and the commercial pressure behind it is unbearable.”

The race to get artificial intelligence to market has raised the risk of a Hindenburg-style disaster that shatters global confidence in the technology, a leading researcher has warned.Michael Wooldridge, a professor of AI at Oxford University, said the danger arose from the immense commercial pressures that technology firms were under to release new AI tools, with companies desperate to win customers before the products’ capabilities and potential flaws are fully understood.

OpenClaw, like many other open-source tools, allows users to connect to different AI models via an application programming interface, or API. Within days of OpenClaw’s release, the team revealed that Kimi’s K2.5 had surpassed Claude Opus and became the most used AI model—by token count, meaning it was handling more total text processed across user prompts and model responses.

Low-cost Chinese AI models forge ahead, even in the US, raising the risks of a US AI bubble Nvidia’s latest earnings report reassured some. But Chinese AI models are fast gaining a following around the world, underlining concerns over an ‘AI bubble’ centered on high-investment, high-cost US models.

One of the largest PC suppliers, Dell, was reported to be planning a price hike that could raise hardware costs by hundreds of dollars. Interestingly, for consumers opting for higher memory configurations, this would now require a significant price increase. Here were the price increases that were reported across a variety of products: $130–$230 increase for Dell Pro and Pro Max notebooks and desktops configured with 32 GB of memory $520–$765 increase for systems configured with 128 GB of memory $55–$135 increase for configurations with a 1 TB SSD $66 increase for AI laptops equipped with an NVIDIA RTX PRO 500 Blackwell GPU (6 GB) $530 increase for AI laptops equipped with an NVIDIA RTX PRO 500 Blackwell GPU (24 GB) Similarly, companies like ASUS and Acer were also reported to be bumping up PC pricing to cope with memory shortages, and according to Acer's Chairman, Jason Chen, the BoM (Bill of Materials) for several products within Acer's portfolio has risen dramatically, leaving no choice but to increase prices to ensure consistent supply. Small-scale manufacturers like Framework are also looking to increase the cost of upgrading RAM on existing configurations, indicating a widespread "price hike" wave approaching gamers.

the humans involved may have simply lost the plot and may not understand what the program is supposed to do, how their intentions were implemented, or how to possibly change it.

AI infrastructure developers cannot wait five years. In many cases, they cannot wait six months, because waiting six months costs billions of dollars of lost opportunities.

we might move again. The point is that we can. We can because we own our prompts, our skills, our databases, our memory architecture, they all live in our bar. None of it lives inside OpenAI or Anthropic. When we moved, we rewired the model layer and everything else stayed put. That’s the whole trick, really. If you control the pieces that make your agents smart, switching the engine underneath is just plumbing.

What if I actually did have dirt on me that an AI could leverage? What could it make me do? How many people have open social media accounts, reused usernames, and no idea that AI could connect those dots to find out things no one knows?

AI Doesn’t Reduce Work—It Intensifies It

I’m going to cure my girlfriend’s brain tumor.